Characterization of Cytosolic Glutathione Peroxidase And

Home , Deiodinase, Glutathione oxidase, GPX1, GPX2 (gene), GPX3, GPX4

Aquatic Toxicology 130–131 (2013) 97–111

Contents lists available at SciVerse ScienceDirect

Aquatic Toxicology

jou rnal homepage: www.elsevier.com/locate/aquatox

Characterization of cytosolic glutathione peroxidase and

phospholipid-hydroperoxide glutathione peroxidase genes in rainbow trout

(Oncorhynchus mykiss) and their modulation by in vitro selenium exposure

a a b a d c a,∗

D. Pacitti , T. Wang , M.M. Page , S.A.M. Martin , J. Sweetman , J. Feldmann , C.J. Secombes

Scottish Fish Immunology Research Centre, Institute of Biological and Environmental Sciences, University of Aberdeen, Aberdeen AB24 2TZ, United Kingdom

Integrative and Environmental Physiology, Institute of Biological and Environmental Sciences, University of Aberdeen, Aberdeen AB24 2TZ, United Kingdom

Trace Element Speciation Laboratory, Department of Chemistry, University of Aberdeen, Aberdeen AB24 3UE, United Kingdom

Alltech Biosciences Centre, Sarney, Summerhill Rd, Dunboyne, Country Meath, Ireland

a r t i c l e i n f o a b s t r a c t

Article history: Selenium (Se) is an oligonutrient with both essential biological functions and recognized harmful effects.

Received 4 July 2012

As the selenocysteine (SeCys) amino acid, selenium is integrated in several Se-containing proteins

Received in revised form

(selenoproteins), many of which are fundamental for cell homeostasis. Nevertheless, selenium may exert

19 December 2012

toxic effects at levels marginally above those required, mainly through the generation of reactive oxygen

Accepted 20 December 2012

species (ROS). The selenium chemical speciation can strongly affect the bioavailability of this metal and

its impact on metabolism, dictating the levels that can be beneﬁcial or detrimental towards an organism.

Keywords:

Glutathione peroxidase (GPxs) is the largest and the most studied selenoprotein family. Cytosolic glu-

Selenium

tathione peroxidase (cGPx, GPx1) and phospholipid hydroperoxide glutathione peroxidase (PHGPx, GPx4)

Sodium selenite

Selenocysteine are widely distributed throughout tissues, and play a pivotal role in regulating the oxidative status in the

Glutathione peroxidase cell. In this study we have cloned GPx1 and GPx4 genes in rainbow trout (Oncorhynchus mykiss). The con-

Salmonids stitutive mRNA expression of these GPx genes was examined in 18 trout tissues and their responsiveness

RTL cell line to Se availability was analysed using a rainbow trout liver cell line (RTL). An inorganic (sodium selenite,

Na2SeO3) and organic (selenocysteine, Cys-Se-Se-Cys) selenocompound have been used as Se sources.

GPx1 activity was also tested to verify the impact of transcript changes on the enzymatic function of these

molecules. To understand if the results obtained from the transcript expression analysis were due to Se

bioavailability or generation of ROS, the cytoxicity of the two selenocompounds was tested by measuring

the impact of Se on cell membrane integrity. Lastly, Se availability was quantiﬁed by mass spectropho-

tometry to determine the amount of Se in the cell culture media, the Se background due to the foetal calf

serum supplement and the contribution from the two selenocompounds used in the treatments.

Three isoforms of genes for both GPx1 (GPx1a, 1b1 and 1b2) and GPx4 (GPx4a1, a2 and b) have been

identiﬁed. The discovery of a third gene encoding for GPx1 and GPx4 hints that salmonids may have the

biggest selenoproteome amongst all vertebrates. Transcripts of GPx4 genes were more highly expressed

in most tissues examined in vivo (except blood, head kidney and spleen), whereas those of the GPx1 genes

were more responsive to selenium exposure in vitro, especially to the organic form. Interestingly, GPx1a

was the most sensitive to selenium availability in non stressful conditions, whereas GPx1b1 and GPx1b2

were highly induced by exposure to selenium levels that had some toxic effects on the cells. Although

the different concentrations tested of the two selenocompounds modulate GPx1 transcript expression to

various degrees, no signiﬁcant change of GPx1 enzymatic activity was detectable. Our results lead us to

conclude that trout GPx1 transcripts expression level may represent a sensitive biomarker for selenium

intake, helping to evaluate if selenium concentration and chemical speciation impact on cell homeostasis.

1. Introduction

Selenium (Se) is an essential trace element, required as an

∗

Corresponding author at: Scottish Fish Immunology Research Centre, Institute

integral part of diverse Se-containing proteins, called seleno-

of Biological and Environmental Sciences (IBES), University of Aberdeen, Tillydrone

proteins (Burk and Hill, 1993). Through its incorporation into

Avenue, Aberdeen, AB24 2TZ, United Kingdom. Tel.: +44 1224 272857;

selenoproteins as the selenocysteine (SeCys) amino acid, selenium

fax: +44 1224 272872.

E-mail address: [email protected] (C.J. Secombes). exerts its biological effects primarily by regulating the antioxidant

98 D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111

system and redox enzyme activities within the cell (McKenzie et al., non-speciﬁcally into proteins instead of methionine, leading to sig-

2002). Moreover, this trace element is involved in gene transcrip- niﬁcant alterations in protein structure and consequently protein

tion and cell signalling cascades, thyroid hormone metabolism, function (Schrauzer, 2000).

immune responses and reproduction (Rayman, 2000; Hefnawy and SeCys is the most relevant selenocompound, as the naturally

Tórtora-Pérez, 2010). However, Se toxicity can be reached by lev- occurring amino acid (the 21st) that provides the catalytic site for

els marginally above those which are required. Among the essential the enzymatic selenoproteins (Stadtman, 1996). SeCys is character-

trace elements, Se presents the narrowest range between essential- istically different from the remaining amino acids as it is encoded

ity and toxicity, and for this reason it is often difﬁcult to establish by one of the three stop codons (UGA) and is inserted into the

[SeCys]

which concentrations are beneﬁcial and which become detrimen- selenoproteins by the tRNA , which has speciﬁc features that

tal towards an organism’s health (NRC, 1980; Foster and Sumar, distinguish it from all other tRNAs (Bock et al., 1991). The UGA

1997). codon is made to encode SeCys by the presence of a cis-acting stem-

Diet and water are the two sources of Se. In the aquatic envi- loop structure, designated the SeCys insertion sequence (SECIS)

ronment, Se can be absorbed through the gills, gut or epidermis; element, present in the 3 un-translated regions (3 -UTRs) in the

however diet is considered the primary source of Se for animals mRNA (Walczak et al., 1996). In addition, other factors are required

and the intestine the principal route of assimilation (Dallinger for the incorporation of SeCys into the mature protein, namely

et al., 1987; Hamilton, 2004; Janz, 2011). The immediate bioavail- SECIS-binding protein 2 (SBP2) (Copeland et al., 2000), and SeCys-

ability of Se depends on its chemical form, which determines the speciﬁc elongation factor (EFsec, also called mSelB) (Fagegaltier

metabolic and toxic potential (Jonnalagadda and Prasada Rao, 1993; et al., 2000). SECIS elements are present in the 3 un-translated

Jackson, 1997; Finley, 2006). Se exists in the environment and in regions (3 -UTRs) of all eukaryotic selenoprotein genes (Berry et al.,

biological systems as both inorganic and organic forms. The inor- 1991). The general structure of the SECIS element includes two

2− 2−

ganic salts, selenite (SeO3 ) and selenate (SeO4 ), contain Se in helices separated by an internal loop, a SECIS core structure located

oxidized forms (Se(IV) and Se(VI) respectively), whereas organic at the base of helix 2 and one or two apical loops; its conﬁguration

forms contain Se in the reduced state (selenide: Se(−II)) (Thomas can be selenoprotein- and species-speciﬁc. The presence of a SECIS

et al., 1990; Birringer et al., 2002). Generally, selenocysteine (SeCys) dictates any in-frame UGA codon within the coding region to serve

and selenomethionine (SeMet) are the most abundant selenocom- as SeCys when a minimal spacing requirement between UGA and

pounds present in the diet (Suzuki, 2005; Dumont et al., 2006). The the SECIS element (51 to 111 nucleotides) is met (Low and Berry,

diverse speciation of selenium suggests that the cellular uptake 1996; Fletcher et al., 2001). A set of selenoproteins in an organism

of this trace element is likely to occur via multiple membrane is known as the selenoproteome. To date 25 selenoprotein genes

transporters (Misra et al., 2012b). Previous studies, focused on the have been indentiﬁed in humans (Kryukov et al., 2003). Recent

mechanisms of intestinal absorption of selenocompounds in mam- studies on the vertebrate selenoproteome showed that bony ﬁsh,

2−

mals, suggest that SeO3 absorption occurs by simple diffusion, with 38 selenoprotein genes identiﬁed in zebraﬁsh, have a larger

2−

whilst SeO4 may be transported by a common transport mech- set of selenoproteins than humans and all terrestrial vertebrates

anism with sulphate (Arduser et al., 1985; Wolffram et al., 1986). (Mariotti et al., 2012).

SeMet appears to be absorbed by an active transport system shared Glutathione peroxidase (GPxs) is the largest and the most stud-

with methionine (McConnell and Cho, 1965; McConnell and Cho, ied selenoprotein family (Flohé and Brigelius-Flohé, 2012). GPxs,

1967; Wolffram et al., 1989), whereas SeCys has been reported to be together with other antioxidant enzymes (i.e. catalase, glutathione

either passively absorbed (McConnell and Cho, 1965) or absorbed reductase and superoxide dismutase), regulate the oxidative status

via a competitive transport with cysteine in brush-border mem- in the cell reducing either free or membrane-bound hydroperox-

branes (Wolffram et al., 1989; Leblondel et al., 2001). Currently, ides (Apel and Hirt, 2004). Proteins within the GPx family differ in

the mechanisms of Se transport in ﬁsh, particularly at the cellu- structure, substrate speciﬁcity and cellular location, with some hav-

lar level, are largely unknown. However, it is believed that the ing a tissue-speciﬁc distribution (Arthur, 2000). Eight GPx isoforms

conversion of selenocompounds into metabolically active forms have currently been characterized in humans, six of which (GPx1,

is the rate–limiting step for Se bioavailability (Contempre et al., GPx2, GPx3, GPx4, GPx6, GPx8) are SeCys-containing proteins, the

1996). In vertebrates, inorganic selenocompounds react sponta- remaining two (GPx5 and GPx7) have a cysteine residue instead

neously with glutathione (GSH) to form selenodi-diglutathione of selenium (Mariotti et al., 2012; Nauser et al., 2012). Among the

(GSSeSG) (Ganther, 1968), which is further converted to hydro- GPx proteins, GPx1 and GPx4 were the ﬁrst discovered and are the

gen selenide (H2Se) (Steve Hsieh and Ganther, 1975). This central best characterized (Flohé et al., 1973; Rotruck et al., 1973; Ursini

selenide pool serves either as the basis for methylated derivates et al., 1982; Toppo et al., 2008). GPx1 (also called cytosolic glu-

that are excreted or as a precursor for further transformation into tathione peroxidase, cGPx) and GPx4 (phospholipid hydroperoxide

SeCys (Ganthert, 1966; Mozier et al., 1988; Ip et al., 1991; Berry glutathione peroxidase, PHGPx) are widely distributed throughout

et al., 1993; Guimarães et al., 1996). Vertebrates are also able tissues and are highly conserved across all species (Margis et al.,

to synthesize SeCys but not SeMet, which is usually metabolised 2008). GPx1, is a homotetramer localised in the cytosol, nucleus

through the trans-sulfuration pathways of the sulphur amino and mitochondria, where it acts to neutralise hydrogen peroxide

acids and then converted into SeCys (Beilstein and Whanger, (H2O2) (Flohé et al., 1973). The kinetic properties of GPx1 imply that

1992). under physiological substrate concentrations, the maximum rate is

The mechanism of Se toxicity has not been clearly elucidated. never reached, i.e. the actual reaction rate is always under the con-

It is hypothesised that it is related to the generation of reactive trol of the amount of enzyme (Wendel, 1981). Contrary to the other

oxygen species (ROS) caused by the depletion of glutathione and members of the GPxs family, GPx4 possesses a monomeric struc-

protein-bound sulphydryl groups that occur during the metabolism ture and hydrophobic surface (Ursini et al., 1985), which underlies

of inorganic selencompounds (Anundi et al., 1984; Ganther, 1986). a unique catalytic activity capable of reducing complex membrane

Also, SeMet can be transformed through an alternative path- bound hydroperoxides and lipid peroxides, such as phospholipid

way involving methylselenol, which is subsequently metabolized and cholesterol hydroperoxides (Thomas et al., 1990; Roveri et al.,

in the GSH-mediated redox cycle with the production of ROS 1994; Schnurr et al., 1996). GPx4 localises primarily to cellular

(Chaudiere et al., 1992). Finally, Se may interfere with sulphur membranes, and its main function is the protection of plasma and

metabolism, which can occur under both toxic and non-toxic condi- lysosomal membranes from ROS-induced damage (Ursini et al.,

tions (Combs and Combs, 1986). Indeed, SeMet can be incorporated 1985).

D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111 99

Table 1

Primers used for cloning and real time PCR for trout GPx genes.

Gene name Primer name Primer sequence (5 →3 ) Application

Glutathione peroxidase 1a F1 GGGTCAGTTGCTGTACAAAAGC 3 -RACE F2 GGTATCCGAAAGTTAGACGTCGTTG 3-RACE

F ATGAAATGGCTGGGAAAATAAAGA qPCR

R TCATCATTCTTACAATTCTCCTGATG qPCR

Glutathione peroxidase 1b1 F1 AGCCCTGTTTAGCAGACAGAAAAAGC 3 -RACE

F2 AACATGTCTGGAAGTGAGTTCTACAACA 3 -RACE

F CAACATGTCTGGAAGTGAGTTCTACAACA qPCR R TTCGTTATTGCAGTTCTCCTGATGTC qPCR

Glutathione peroxidase 1b2 F1 AGAGAGAAGCGGGGAGTGACA 3 -RACE

F2 ATGGCTGTATGTAAGATGTTCTACGACC 3 -RACE

F ACCAGGCAAATGGCTGTATGTAAGAT qPCR

R CTTCGTTCTTGCAGTTCTCCTGATG qPCR

Glutathione peroxidase 4a1 F1 GATTTGAGTACGCAGTACGTTTCCCA 3 -RACE

F2 CCAACAAACCCTGCAGACGG 3 -RACE

F GTAAGATTGACGTGAACGGGGATAA qPCR

R GAAATTCCACTTGATGCCATTTCC qPCR

Glutathione peroxidase 4a2 F1 GACTTCTCTGCAAAGGATATTGATGG 3 -RACE

F2 ACCCCAGTCAACTACTCTCAGTTTGC 3-RACE

F CCTCGAGAAATACAGGGGCGA qPCR

R TGAAATTCCACTTGATGCCATTTC qPCR

Glutathione peroxidase 4b F1 CTACTTAAACAGCTTGGTCACAGGATG 3 -RACE

F2 GGATCGTAACGCGTGGGTTAG 3 -RACE

F AGGTGGAAATCAAGGAGTTTGTCAAAC qPCR R CTTTTGTGAAATTCCATTTGATGTTATTTCC qPCR

␣

Elongation factor-1 F CAAGGATATCCGTCGTCGTGGCA qPCR

R ACAGCGAAACGACCAAGAG qPCR

In this study we have ﬁrst cloned and characterized the homo- in the cell culture media, the Se background due to the foetal calf

logues of mammalian GPx1 and GPx4 genes in rainbow trout serum supplement and the contribution from the two selenocom-

(Oncorhynchus mykiss). The transcript expression of these GPx pounds used in the treatments.

genes was then examined in different trout tissues. Next, we estab-

lished the rainbow trout liver cell line (RTL) as a model to analyse 2. Materials and methods

the responsiveness of GPx genes to Se, as a means to study Se

effects in a controlled system prior to future mechanistic stud- 2.1. Cloning and sequencing of trout GPx1 and GPx4

ies in whole animal experiments. The liver is known to be the

principal tissue involved in Se metabolism and RTL are a well BLAST search (Altschul et al., 1997) at NCBI

established model to determine the effect of various exogenous (http://blast.ncbi.nlm.nih.gov/Blast.cgi) of the salmonid EST

substances on ﬁsh cells (Lee et al., 1993). To verify whether different database using the zebraﬁsh GPx1a (Acc. No. NM 001007281),

effects on transcript expression of GPx gene are induced by inor- GPx1b (Acc. No. NM 001004634), GPx4a (Acc. No. NM 001007282)

ganic and organic selenocompounds, sodium selenite (Na2SeO3) and GPx4b (Acc. No. NM 001030070) as baits, identiﬁed a salmon

and selenocysteine (Cys-Se-Se-Cys) were both used as Se sources partial sequence encoding for GPx4 (Acc. No. BT044014), a par-

at different concentrations. Na2SeO3 is the most common inor- tial trout sequence encoding for GPx1 (Acc. No. AF281338) and

ganic Se form and a commonly used compound in the study of the over 30 expressed sequence tags (ESTs) from an Oncorhynchus

effects of this trace element both in vitro and in vivo (Spallholz, mykiss cDNA library. The ESTs were assembled with the AlignIR

1997; Valdiglesias et al., 2010). Cys-Se-Se-Cys is a stable form of programme (Li-COR, Inc.) and 3 contigs for each of GPx1 and

SeCys, and consists of two oxidized selenocysteine residues that GPx4 were detectable that contained the 5 -untranslated region

form a diselenocompound (Huber and Criddle, 1967). Lately, Cys- (5 -UTR) and part of the open reading frame (ORF). Primers F1

Se-Se-Cys has received more attention due to its potential use as and F2 were designed at the 5 -UTR (Table 1) of each contig, to

a food/feed additive and its pharmacological applications. More- obtain the full ORF and the 3 -untranslated region (3 -UTR). A

over, recent studies have shown that Cys-Se-Se-Cys is the principal 3 -RACE PCR was carried out using SMART cDNA from liver and

metabolite in the liver of ﬁsh feed organic selenium compounds, as spleen as described before (Wang and Secombes, 2003; Wang

SeMet (Misra et al., 2012a). However, to date only few investiga- et al., 2008). The nested RACE products were ligated into pGEM-T

tions have been conducted on the balance between the beneﬁcial easy vector (Promega) and transformed into E. coli TAM competent

and toxic effects of this diselenocompound (El-Sayed et al., 2006; cells (ActiveMotif), following the manufacturer’s protocol. Positive

Santhosh Kumar et al., 2009). GPx1 genes were found to be the most clones were screened by standard colony PCR and cultured at

◦

sensitive to selenium exposure, thus we also analysed GPx1 enzy- 37 C overnight in a shaking incubator (200 rpm) before plasmid

matic activity after Se treatment. Due to the dual nature of Se, as isolation. Plasmid DNA from at least four clones per PCR was

an essential component of antioxidant enzymes and pro-oxidant extracted using a Qiagen minipreparation kit and sequenced by

compound, the cytoxicity of the two selenocompounds was tested MWG Biotech (Germany).

to elucidate if the effects seen at the transcript expression level

could be due to Se bioavailability or Se toxicity. Hence cell viabil- 2.2. Sequence analysis

ity was determined after treatment with different concentrations of

the two selenocompounds studied. Lastly, Se availability was quan- The nucleotide sequences generated were assembled and

tiﬁed by mass spectrophotometry to determine the amount of Se analyzed with the AlignIR program (version 2.0; LI-COR). The

100 D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111

intron positions were predicted using the Spidey program at the 1 killing. RNA was subsequently extracted as described in Section

NCBI (http://www.ncbi.nlm.nih.gov/IEB/Research/Ostell/Spidey/). 2.3, and the transcript expression of GPx1 and GPx4 gene isoforms

The SECIS elements were predicted using SECISearch (version was analyzed by real time PCR, as described in Section 2.4.

2.19; http://genome.unl.edu/SECISearch.html) (Kryukov et al.,

2003). Each preliminary cDNA sequence was translated using 2.6. Transcript expression of trout GPx1 and GPx4 in RTL cells

the ExPASy translate tool (http://web.expasy.org/translate/) after Na2SeO3 and Cys-Se-Se-Cys treatment

(Gasteiger et al., 2005). To obtain the similarity and identity

of the sequences the software MatGat (version 2.02) was used RTL cells were used to assess the transcript expression of GPx1

(Campanella, 2003). Multiple sequence alignments were gener- and GPx4 genes in a characterized trout cell line (Lee et al., 1993).

ated using ClustalW (http://www.ebi.ac.uk/Tools/msa/clustalw2/) The cells were maintained in Leibovitz medium (L-15, Gibco) con-

(Chenna, 2003) and shaded using BOXSHADE (version 3.21; taining 10% foetal bovine serum (FBS) (Sigma), 100 U/ml penicillin

◦

http://www.ch.embnet.org/software/BOX form.html). Based on a and 100 ␮g/ml streptomycin (P/S) at 20 C. Four days before stimu-

ClustalW multiple alignment phylogenetic trees were created by lation the cells were trypsinised, washed twice by centrifugation at

the neighbour-joining method using the MEGA program (version 200 × g for 5 min and passaged at a concentration of 1 × 10 cells/ml

5.0) (Tamura et al., 2011), using Pairwise Deletion of gaps/missing into 12-well culture plates (Millipore), with 2 ml of fresh medium

◦

data and JTT amino acid matrix, and bootstrapped 10,000 times. and incubated at 20 C. After 2 days the medium was replaced with

L-15 containing 1% FBS and P/S. One day later, the medium was

2.3. RNA extraction and cDNA synthesis replaced with fresh medium containing either an inorganic (sodium

selenite, Na2SeO3) or an organic (Selenocysteine L-stereoisomer,

Total RNA was extracted from 100 mg of tissue that had been Cys-Se-Se-Cys) selenocompound, both purchased from Sigma-

homogenised in TRIZol (Invitrogen) using tungsten carbide beads Aldrich. The cells were incubated with six different concentrations

(3 mm, Qiagen) and the TissueLyser II system (Qiagen) following of the two selenocompounds: 0 (control), 10 nM, 100 nM, 1 ␮M,

the manufacturer’s instructions. The RNA pellet was washed twice 10 ␮M and 50 ␮M. These concentrations were deemed optimal for

with 70% ethanol, air dried and resuspended in RNase free H2O. The stimulation of RTL cells from our previous studies (data not shown).

concentration and purity was determined by spectrophotometry After 24 h, the treatments were terminated by dissolving the cells in

◦

(Nanodrop). The RNA was stored at −80 C until required for cDNA TRIZol. Total RNA was then extracted from the cells and cDNA syn-

synthesis. cDNA was synthesized from total RNA: 2 ␮g of total RNA thesized as described in Section 2.3. The dose-dependent effects

◦

was denatured (70 C, 3 min) in the presence of 1 ␮l (500 ␮g/ml) of Se on the transcript expression of GPx1 and GPx4 genes were

of an adaptor-dT17 (Table 1) and RNase free water in a volume of analyzed by real time PCR, as described in Section 2.4.

11 ␮l, then the RNA was cooled on ice. The ﬁrst strand cDNA was

synthesized from the total RNA using 1 ␮l of RevertAid reverse 2.7. GPx1 enzymatic activity in RTL cells after Na2SeO3 and

transcriptase (10,000 U, Fermentas) in the presence of 5 ␮l of 5× Cys-Se-Se-Cys treatment

Reaction Buffer, 1 ␮l of dNTP (deoxynucleoside triphosphate mix

25 mM each) (Bioline), made up to a ﬁnal volume of 25 ␮l with Cells were treated with the two selenocompounds investigated

◦

water and incubated at 42 C for 2 h. as described in Section 2.6, with the exception that they were

passaged into 6-well culture plate (Millipore), with 3 ml of fresh

2.4. Real time reverse transcriptase PCR (qPCR) medium.

After 24 h of treatment, the cells were rinsed once with ice cold

Real time PCR was performed with a LightCycler 480 (Roche) to phosphate buffered saline followed by incubation for 10 min in ice

quantify the transcript expression of the GPx genes and the com- cold lysis buffer, containing 10 mM Tris (pH 8.0), 150 mM NaCl,

mon reference gene using the primers given in Table 1. The primers 2 mM EDTA, 2 mM dithiothreitol (DTT) and 40% glycerol. A pro-

employed for real time PCR were designed with at least one primer tease inhibitor cocktail tablet (Roche) was added to the lysis buffer

across a predicted intron and pre-tested to ensure that each primer immediately before use, following the manufacturer’s protocol.

pair could not amplify genomic DNA using the real time PCR pro- Cells were scraped from the plates and cell lysates were centrifuged

◦

tocols. The real time PCRs were performed in duplicate for each (Peqlab PerfectSpin) at 13,000 × g, for 15 min, at 4 C. Protein con-

sample, along with a 10-fold serial dilution of references consisting centration was determined using the Bradford technique (Bradford,

of an equimolar mix of puriﬁed PCR products of each gene ampliﬁed 1976) with a BioRad protein kit. Whole cell lysates were stored at

− ◦ from cDNA. The transcript level was calculated using the quantita- 80 C.

◦

tive ﬁt points method in the integrated LightCycler 480 software. GPx1 activity was determined at 30 C, with values read in a

The relative transcript expression level of each gene in different tis- Spectramax Plus384 (Molecular Devices). Conditions for the GPx

sues was expressed as arbitrary units that were calculated from the activity assay were: 50 mM potassium phosphate buffer (pH 7.0),

serial dilution of references run in the same 96-well plate and nor- 0.4 mM EDTA, 0.15 mM ␤-NADPH, 1 unit glutathione reductase,

malized against the transcript expression level of EF-1˛ gene. Fold and 1 mM reduced glutathione. Background changes in absorbance

changes were also calculated as the average expression level of each were measured with the above components as well as with the

treatment group divided by that of the corresponding controls. addition of 10 ␮l of cell lysate. The reaction was initiated by the

addition of 0.007% H2O2 and the change in absorbance was mea-

2.5. Tissue distribution of the transcript expression of the trout sured at 340 nm for 2 min.

GPx1 and GPx4 isoforms

2.8. Membrane integrity of RTL cells after Na2SeO3 and

Juvenile rainbow trout (Oncorhynchus mykiss), weighing Cys-Se-Se-Cys treatment

approximately 100 g, were maintained in 250 l freshwater tanks at

the aquarium facilities in the School of Biological Sciences, Univer- Cells were seeded into 96-well plates at a concentration of

× 5 sity of Aberdeen. The fish were fed ad libitum. Tail fin, adipose fin, 2 10 cells/ml, with 200 ␮l of medium per well, and incubated

◦

scales, skin, brain, thymus, gills, head kidney, spleen, blood, heart, at 20 C. In all cases the ﬁrst column of wells was left without cells

ovary, liver, eosophagus, stomach, pyloric caeca, mid intestine and (blank). After 48 h the medium was replaced with L-15 contain-

distal intestine were collected from six healthy ﬁsh after schedule ing 1% FBS and P/S. After 24 h, the medium was replaced with fresh

D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111 101

medium containing Na2SeO3 and Cys-Se-Se-Cys, at a concentration Table S1). However, since these molecules grouped in phylogenetic

of 10 nM, 100 nM, 1 ␮M, 10 ␮M and 50 ␮M. The cytotoxic effect of tree analysis (discussed later) with other ﬁsh molecules previously

the selenocompounds was examined at 24 h, by analysing the lac- identiﬁed and termed GPx1’s, we named the trout sequences as

tate dehydrogenase (LDH) activity released from damaged cells into GPx1a, 1b1 and 1b2, respectively. The translations of the other three

the supernatant (Korzeniewski and Callewaert, 1983). To eliminate cDNA sequences (Acc. Nos. HE687024-6) shared high identities

effects of potential damaged or detached cells after treatment, the (59.5–78%) to GPx4 from zebraﬁsh and mammals (Supplementary

cell culture was ﬁrst centrifuged (250 × g) for 10 min at room tem- Table S1) and were designated as GPx4a1, 4a2 and 4b. The designa-

perature at the end of the incubation period. 100 ␮l of supernatants tion of the trout sequences was further studied by phylogenetic tree

were then transferred into a new 96-well plate and the LDH release analysis (Fig. 1) using the MEGA 5 program (Tamura et al., 2011).

determined using a cytotoxicity detection kit (Roche, Germany) The Se-dependent GPx proteins (GPx1, GPx2, GPx3 and GPx4) from

according to the manufacturer’s guidelines. Cells exposed to 1% Tri- ﬁsh and selected mammals were employed for this analysis as GPx6

ton X-100 were used as the “high” control (100% lysis), while cells and GPx8 have not yet been characterized in ﬁsh. MEGA 5 does not

exposed to culture medium only acted as the “low” control. The recognize the SeCys residue. Thus it was substituted with a cys-

changes in absorbance were recorded in a spectrophotometer (as teine residue, which normally replaces the SeCys position in the

above) at 490 nm, using 600 nm as a reference wavelength. Se-independent GPxs, and in all the non-SeCys-containing homo-

logues of selenoproteins across the different phyla. As shown in

2.9. Chemical analysis Fig. 1, three independent clades are apparent, for GPx1/2, GPx3 and

GPx4. In the GPx1/2 clade, the trout GPx1a grouped with zebraﬁsh

The total selenium concentration in the in vitro model used was GPx1a and fugu GPx1a to form a branch; and the trout GPx1b1

measured by Inductively Coupled Plasma Mass Spectrometry (ICP- and 1b2 grouped with zebraﬁsh GPx1b and fugu GPx1b forming

MS, Thermo), to ensure the absence of interference of Se contained another branch. All the ﬁsh GPx1-like molecules grouped together

in the FCS and to verify Se availability in the cell cultures treated to form a sub-clade separate from both the mammalian GPx1 and

with Na2SeO3 and Cys-Se-Se-Cys. The concentration of the two GPx2 sub-clades. In the GPx4 clade of the phylogenetic tree, two

77 78

Se isotopes ( Se and Se) was measured in duplicate in undi- sub-groups of ﬁsh GPx4, containing respectively trout GPx4a1 and

luted FCS, L-15 medium, L-15 medium containing 1% FCS (control), 4a2 in one group and trout GPx4b in the other, grouped with mam-

and in the medium of the cell cultures treated with the different malian GPx4 with high bootstrap support (100%). The tree topology

concentrations of Na2SeO3 and Cys-Se-Se-Cys (as listed in Section suggests that the ﬁsh GPx1 molecules are not strictly orthologues

2.6). De-ionized water (18 mohm, Mellipore Element System) was of mammalian GPx1 or GPx2, and are more likely related to an

used for sample dilution (1:10) and standard preparation. Nitric ancestral molecule that gave rise to GPx1 and GPx2 later in ver-

acid (ultra pure, 70%, Australian Chemical Reagents), and analytical tebrate evolution. In contrast, the ﬁsh GPx4 molecules appear to be

grade hydrogen peroxide (30%, Analar, BDH) were used for sam- orthologues of mammalian GPx4, although multiple paralogs are

ple digestion. Duplicate samples were analyzed with calibration, present.

␮

reagent blanks and reference material (human urine, 68 g/l of Se, Each cDNA sequence of the trout GPxs obtained in this study

Serenorm) to check quality assurance and quality control at the contains an in-frame stop codon “TGA” in the ORF encoding for the

beginning and end of each ICP-MS run. selenocysteine amino acid (SeCys, U; Supplementary Fig. S1), pos-

sibly with the exception of GPx4b that may have two additional

2.10. Statistical analysis selenocysteine codons (Supplementary Fig. S1F). The TGA codon is

made to encode selenocysteine by the presence of a SECIS element

Real-time quantitative PCR measurements were analyzed using in the 3 -UTR of the mRNA (Hatﬁeld, 2002). Indeed, a SECIS ele-

the SPSS package 18.0 (SPSS Inc. Chicago, Illinois) as described ment region can be identiﬁed in each trout cDNA sequence within

previously (Wang et al., 2011). To improve the normality of data its 3 -UTR using the SECISearch programme (Fig. 2). The SECIS ele-

distribution, a log2 transformation of the arbitrary units after ments in trout GPx1/2 genes contain the non-Watson-Crick base

normalization to the transcript expression level of EF-1 was per- pairs 5 TGAA3 :5 TGAT3 , whilst 5 TGAC3 :5 TGAT3 is present in

formed. One way-analysis of variance (ANOVA) and the Least the trout GPx4 isoforms. All the GPx isoforms possess two apical

Signiﬁcant Difference (LSD) post hoc test were used to analyze the loops and hence they belong to the type II group of selenoproteins

expression data and the viability response in RTL cells after stimu- (Latrèche, 2009).

lation, with p < 0.05 between treatment groups and control groups The predicted polypeptides of the trout GPxs were aligned

considered signiﬁcant. Paired-sample t-tests were employed to with GPx proteins from ﬁsh and mammals. As shown in Fig. 3,

compare the expression levels across different tissues and differ- the “L(V/I)VN(VT)ASx(U)G(L/F)TxxxYxxLxxL” motif surrounding

ent isoforms in the same tissue (as six sets of samples from six the reactive selenocysteine residue was well conserved. Simi-

individual ﬁsh were used in this study), and the quantiﬁed levels larly, the “G(L/F)x(V/I)L(G/A)FPCNQFxxQEP” and “WNFxxKFL(V/I)”

77 78

of the two selenium isotopes ( Se and Se) measured by ICP-MS sequences that surround the glutamine, tryptophan and asparagine

in duplicate in undiluted FCS, L-15 medium, L-15 medium contain- residues also involved in the catalytic site were conserved.

ing 1% FCS (control), and in the medium of the cell cultures treated

with the different concentrations of selenocompounds. 3.2. Tissue distribution of the expression of GPx1 and GPx4 isoforms

3. Results

The constitutive transcript expression of the GPx1 and GPx4

3.1. Sequence analysis of trout GPx1 and GPx4 gene isoforms gene isoforms was comparatively studied in 18 tissues from healthy

ﬁsh, by qPCR (Fig. 4).

Six cDNA sequences, each with a complete open reading frame, GPx1b2 showed the highest expression level in the head kidney,

5 -UTR and a 3 -UTR, including a polyadenylation signal and a poly spleen and in the blood, although only in the blood was the expres-

A tail, have been cloned in this study (Supplementary Fig. S1). The sion signiﬁcantly higher compared to the other two gene isoforms.

sequence features are summarized in Table 2. The translations of The expression of GPx1b1 was signiﬁcantly higher in the different

three of the cDNA sequences (Acc. Nos. HE687021-3) shared similar portions of the gastrointestinal tract (eosophagus, stomach, pyloric

identities to mammalian GPx1 and GPx2 (60.1–66%, Supplementary cecae, mid intestine). The expression of GPx1a was comparable or

102 D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111

Table 2

Summary of trout GPx1 and GPx4 isoform features.

Gene name Length (bp) 5 -UTR (bp) ORF (bp) 3 -UTR (bp) SeCys position Poly(A) tail position Amino acid (aa) Acc. No.

Glutathione peroxidase 1a 1062 70 573 419 188–190 1020–1025 190 HE687021

Glutathione peroxidase 1b1 1044 35 570 439 147–149 995–1000 189 HE687022

Glutathione peroxidase 1b2 1061 34 573 454 149–151 1013–1018 190 HE687023

Glutathione peroxidase 4a1 1014 69 603 342 295–297 964–969 200 HE687024

Glutathione peroxidase 4a2 966 61 603 302 287–289 917–922 200 HE687025

Glutathione peroxidase 4b 1038 27 576 435 226–228 986–992 191 HE687026

lower compared to the expression of the other two gene isoforms expression of GPx4 isoforms in the eosophagus was the highest

in all tissues except scales (Fig. 4A). detected, followed by stomach, pyloric cecae and gills.

The transcript expression of GPx4a1 gene was the highest

detected, and signiﬁcantly so compared to the other two iso-

forms, in the gastro-intestinal tract, tail ﬁn, scales and brain. 3.3. Se availability in vitro

GPx4a1 showed the highest transcript expression also in com-

parison to GPx1 isoforms, especially in the ﬁnal portions of the The cells were exposed to an organic (Cys-Se-Se-Cys) and

gastro-intestinal tract (pyloric cecae, mid intestine and distal intes- inorganic (Na2SeO3) selenium compound at ﬁve different concen-

␮

tine). The transcript expression of GPx4a2 was signiﬁcantly lower in trations ranging from 10 nM to 50 M, for 24 h. Total selenium

all tissues, relative to the other two isoforms, as seen in the gastro- concentration of the media in the in vitro model was determined by

intestinal tract, tail ﬁn, scales and brain. GPx4a2 gene showed the using ICP-MS before the start of the incubation period (Table 3). Two

77 78

lowest transcript expression level also in comparison to the GPx1 isotopes ( Se and Se) were measured and the resulting concen-

isoforms. In contrast, the transcript expression of GPx4b was sim- trations for total selenium showed no signiﬁcant differences, hence

ilar to that of GPx4a1, except in the gastro-intestinal tract, tail ﬁn, no interference was recorded. The level of Se in the L-15 medium

scales and brain (Fig. 4B). itself, L-15 medium containing 1% FCS, and the medium contain-

The ratio of the average level of GPx1 and GPx4 transcript ing 10 nM of Na2SeO3 and Cys-Se-Se-Cys was below the limit of

expression in different tissues (Fig. 4) showed a higher constitutive quantiﬁcation (ten times the standard deviation of the blank level

transcript expression of the GPx4 isoforms in all tissue examined, – 1 ␮g Se/L). Since Cys-Se-Se-Cys is a diselenocompound, the level

with the exception of head kidney, spleen and blood. The transcript of selenium into the organic Se treatment was 2.17 ± 0.35 higher

61 Trout GPx1a

76 Fugu GPx1a ACR20471

86 Zebrafish GPx1a NP_001007282 89 Trout GPx1 b2

Trout GPx1b1 GPx1/2

76 Fugu GPx1b ACR20472

99 46 Tuna GPx1 ABO38817

46 Zebrafis hGPx1b NP_001004634

48 Human GPx1 P07203

99 Mouse GPx1 P11352

Bovine GPx1 P00 435

52 Mou se GPx2 Q9JHC0

100 Human GPx2 P18283

65 PigGPx2 Q09HS4

GPx3

Zebrafish GPx3 NP_001131027

Human GPx3 P22352

100 Mouse GPx3 P46412

47 Bovin GPx3 P37141

59 Human GPx4 P36969

100

Mou se GPx4 O70325

Bovin GPx4 Q9N2J2

100 Trou t GPx4a1

GPx4

83 Trout GPx4a 2

100 Tuna GPx4 ABO38818

41 Carp GPx4a ACR33821

84 Zebrafish GPx4a NP_001007283

100 Trout GPx4 b

Salmon GPx4 ACH86324

76 Carp GPx4b ACR33822

0.1 99 Zebrafish GPx4b NP_001025241

Fig. 1. Phylogenetic tree showing the relationship of the predicted trout GPx1 and GPx4 amino acid sequences with other known members of the Se-dependent GPx families

from ﬁsh and mammals. The tree was constructed using ClustalW and MEGA (V.5.0) and inferred using the neighbour-joining method. Node values represent percent

bootstrap conﬁdence derived from 10,000 replicates. The common species name and the name of the molecule is followed by the accession number. The sequences reported

in this report are shaded and without an accession number.

D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111 103

Fig. 2. SECIS elements and multiple alignment of the 3 -UTR of trout GPx1 (A) and GPx4 (B) isoforms. The SECIS elements were generated with SECISearch (V.2.19) and

visualized with RNAnice. The multiple alignment of the 3 -UTR sequences was produced using ClustalW2. The stop codons are underlined, the nucleotides in the quartet and

apical loop are boxed and conserved nucleotides are denoted with an asterisk.

Table 3

than the level in the inorganic Se treatment, at every concentration

Quantiﬁcation of the selenium content of FCS, L-15 medium, L-15 medium plus 1% tested.

FCS (control) and the medium plus different selenium treatments. The concentration

of selenium was quantiﬁed by ICP-MS in each sample. Data are presented as the

77 78

mean and SEM of the amount of two selenium isotopes ( Se and Se) measured

in duplicate, and the p value of a paired-sample t-test comparing them. <1 ␮g/L

3.4. Response of trout GPx1 and GPx4 in RTL cells exposed to

represents data below the detectable level of this method.

Na2SeO3 and Cys-Se-Se-Cys

Samples Selenium measured (␮g/L)

The transcript expression of trout GPx1 and GPx4 was examined

Average SEM p-value

in the trout RTL cell line (Fig. 5). All the GPx1 isoforms had a higher

L-15 <1 – –

sensitivity to the organic Se, and were all signiﬁcantly transcrip-

FCS 11.15 0.29 0.42

tionally induced at the lower concentrations (10 nM and 100 nM)

L-15 + FCS 1% >1 – –

compared to the control and were more highly induced compared

L-15 + FCS 1% + Na2SeO3 10 nM >1 – –

to the inorganic Se treatment. In particular, GPx1a appeared to

L-15+ FCS 1% + Cys-Se-Se-Cys 10 nM >1 – –

be the most responsive gene at the lower (10 nM and 100 nM)

L-15 + FCS 1% + Na2SeO3 100 nM 5.40 0.67 0.21

and intermediated (1 ␮M) concentrations, especially within the

L-15 + FCS 1% + Cys-Se-Se-Cys 100 nM 14.04 0.13 4.30

cells treated with organic Se. GPx1b1 and GPx1b2 induction was

L-15 + FCS 1% + Na2SeO3 1 ␮M 71.73 0.46 0.49

signiﬁcantly higher in the cells treated with organic Se at every con-

L-15 + FCS 1% + Cys-Se-Se-Cys 1 ␮M 137.77 6.83 0.97

centration used, relative to that seen with the inorganic form, and

␮

L-15 + FCS 1% + Na2SeO3 10 M 749.74 6.46 1.00 notably a signiﬁcantly higher transcriptional up-regulation of these

L-15 + FCS 1% + Cys-Se-Se-Cys 10 ␮M 1734.78 47.71 0.99

two genes was seen in the cells exposed to 50 ␮M of both the com-

L-15 + FCS 1% + Na2SeO3 50 ␮M 4005.65 98.31 0.72 pounds tested in contrast to the results with GPx1a. GPx1b2 had the

␮

L-15 + FCS 1% + Cys-Se-Se-Cys 50 M 7403.90 295.98 0.73

highest induction detected of the GPx1 isoforms, at concentrations

of 10 ␮M and 50 ␮M of organic Se.

104 D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111

Sec

TroutGPx1a 1 ------MAGKIKKFYDFSAKLLSG-DLLHFSSLKDKVVLIENVASLUGTTTRDYTQMNELHSQYSEKGLVVLGVPCNQ

ZebrafishGPx1a 1 ------MAGTMKKFYDLSAKLLSG-DLLNFSSLKGKVVLIENVASLUGTTVRDYTQMNELHSRYADQGLVVLGAPCNQ

FuguGPx1 1 ------MAGSVKKFYELTAKLLSG-EVLSFSALRGKVVLIENVASLUGTTARDYTQMNELHGRYAAKGLVILGVPCNQ

TroutGPx1b1 1 ------MSG-SE-FYNMTAKLISG-DFFKFSSLGGKVVLIENVASLUGTTTRDYTQMNELHERYADKGLVILGVPCNQ

TroutGPx1b2 1 ------MAV-CKMFYDLTAKLLTG-ELFNFSSLQGKVVLIENVASLUGTTTRDYTQMNELHERYADKGLVILGVPCNQ

ZebrafishGpx1b 1 ------MAG-IKSFYDITAKTLTG-EEFKFSSLQGKVVLIENVASLUGTTSRDYTQMNELHERFAEKGLVVLGVPCNQ

FuguGPx2 1 ------ME---TQFYDLTAKLLTG-EAFNFSSLQGKVVLIENVASLUGTTSRDYTQMNELHERYAGQGLVILGVPCNQ

MouseGPx1 1 MCAARLSAAAQS--TVYAFSARPLTGGEPVSLGSLRGKVLLIENVASLUGTTIRDYTEMNDLQKRLGPRGLVVLGFPCNQ

HumanGPx1 1 MCAARLAAAAAAAQSVYAFSARPLAGGEPVSLGSLRGKVLLIENVASLUGTTVRDYTQMNELQRRLGPRGLVVLGFPCNQ

MouseGPx2 1 ------MAYIAKSFYDLSAIGLDG-EKIDFNTFRGRAVLIENVASLUGTTTRDYNQLNELQCRFPRR-LVVLGFPCNQ

HumanGPx2 1 ------MAFIAKSFYDLSAISLDG-EKVDFNTFRGRAVLIENVASLUGTTTRDFTQLNELQCRFPRR-LVVLGFPCNQ

TroutGPx1a 72 FGHQENCKNDEILRSLKYIRPGNGFEPKFPLFEKMDVNGKDAHPLFVYLKDKLPFPSDDSMALMSDPKFIMWSPVCRNDV

ZebrafishGPx1a 72 FGHQENCKNEEILQSLKYVRPGNGFEPKFQILEKLEVNGENAHPLFAFLKEKLPQPSDDPVSLMGDPKFIIWSPVCRNDI

FuguGPx1a 72 FGHQENCKNDEILNSLKYVRPGGGFEPKFQLLEKVDVNGKNAHPLFVYLKEKLPFPSDNSMALMADPKFIMWSPVNRNDV

TroutGPx1b1 70 FGHQENCNNEEILMSLKYVRPGNGFEPNFQLLEKVDVNGKHAHPLFVYLKEKLQFPSDDPMALMNDPKCIIWSPVGRNDI

TroutGPx1b2 71 FGHQENCKNEEILMSLKYVRPGNGFEPKFQLLEKVDVNGKDAHPLFVYLKDKLPFPSDEPMALMNDPKCIIWSPVCRTDI

ZebrafishGpx1b 71 FGYQENCTNEEILLSLKYVRPGNGYEPKFQLLEKVDVNGKNAHPLFTFLKEKLPFPSDEPMPFMSDPKFIIWSPVCRNDI

FuguGPx1b 69 FGHQENCKNEEILSSLKYVRPGNGFEPKFQLFEKVDVNVKDAHPLFQFLREKLPFPSDDPTALMSDPKLIIWSPVCRNDV

MouseGPx1 79 FGHQENGKNEEILNSLKYVRPGGGFEPNFTLFEKCEVNGEKAHPLFTFLRNALPTPSDDPTALMTDPKYIIWSPVCRNDI

HumanGPx1 81 FGHQENAKNEEILNSLKYVRPGGGFEPNFMLFEKCEVNGAGAHPLFAFLREALPAPSDDATALMTDPKLITWSPVCRNDV

MouseGPx2 71 FGHQENCQNEEILNSLKYVRPGGGYQPTFSLTQKCDVNGQNEHPVFAYLKDKLPYPYDDPFSLMTDPKLIIWSPVRRSDV

HumanGPx2 71 FGHQENCQNEEILNSLKYVRPGGGYQPTFTLVQKCEVNGQNEHPVFAYLKDKLPYPYDDPFSLMTDPKLIIWSPVRRSDV

TroutGPx1a 152 SWNFEKFLVSPDGDPYKRYSRRFLTSDIEADIKELLN-VK---

ZebrafishGPx1a 152 SWNFEKFLIGPDGEPFKRYSRRFLTIDIDADIKELLKRTK---

FuguGPx1 152 SWNFEKFLIGPDGEPYKRYSRSFLTIDIEADIQELLKRVK---

TroutGPx1b1 150 AWNFEKFLIGPDGEPFKRYSRRFPTIDIEGDIKKLLNTAN---

TroutGPx1b2 151 AWNFEKFLIGPAGEPFKRYGRRFLTSNIEGDIKELLNTAN---

ZebrafishGpx1b 151 AWNFEKFLIGSDGVPFKRYSRRFLTSGIDGDIKKLLSIPK---

FuguGPx2 149 SWNFEKFLIGPDGVPFKRYSRKFLTSSIEGDIKKLLSQAP---

MouseGPx1 159 AWNFEKFLVGPDGVPVRRYSRRFRTIDIEPDIETLLSQQSGNS

HumanGPx1 161 AWNFEKFLVGPDGVPLRRYSRRFQTIDIEPDIEALLSQGPSCA

MouseGPx2 151 SWNFEKFLIGPEGEPFRRYSRSFQTINIEPDIKRLLKVAI---

HumanGPx2 151 AWNFEKFLIGPEGEPFRRYSRTFPTINIEPDIKRLLKVAI--- Sec

TroutG Px4a1 1 ----MSLWQIKRVLVFGTLASCGLVLVIGLYLFTMSAPTEDWQTATSIYDFSAKDIDGNEVSLEKYRGDVVIIVNVASKU

TroutG Px4a2 1 ----MSLWQIKRVLVFGTLASCGLVLVIGLYLFTMSAPTEDWQTASSIYDFSAKDIDGNEVSLEKYRGDVVIIVNVASKU

Zebraf ishGPx4a 1 ------MRFLGSAVVFSLVLQTMSAQLEDWQTAKSIYEFTATDIDGNEVSLEKYRGKVVIITNVASKU

TroutG Px4b 1 ------MWIVTRGLVFGLLGSSG------VGRAMCAQVGDWKTAKYIYEFGAKDIDGEDVALEKYRGFVCIITNVASKU

Salmon GPx4 1 ------SRR------VSRAMCAQVGDWKTAKYIYEFGAKDIDGEDVSLEKYRGFVCVITNVASKU

Zebraf ishGPx4b 1 ------MWLFQRALLVGAVGSKS------FARAMCAQANDWQSAKSIYEFSAIDIDGNDVSLEKYRGYVCIITNVASKU

MouseG Px4 1 MSWGRLSRLLKPALLCGALAAPG------LAGTMCASRDDWRCARSMHEFSAKDIDGHMVCLDKYRGFVCIVTNVASQU

HumanG Px4 1 MSLGRLCRLLKPALLCGALAAPG------LAGTMCASRDDWRCARSMHEFSAKDIDGHMVNLDKYRGFVCIVTNVASQU

TroutG Px4a1 77 GKTPVNYSQFAEMHAKYAEKGLRILAFPSNQFGSQEPGTEAQIKDFAKSYNAEFPMFSKIDVNGDNAHPLWKWLKEQPNG

TroutG Px4a2 77 SKTPVNYSQFAEMHSKYAEKGLRILAFPSNQFGRQEPGTESQIKNFAKSYNADFPMFSKINVNGPNAHPVWKWLKEQPKG

Zebraf ishGPx4a 63 GKTPVNYSQFAEMHAKYSERGLRILAFPSNQFGRQEPGTNSQIKEFAKSYNAEFDMFSKIDVNGDGAHPLWKWLKDQPNG

TroutG Px4b 68 GKTRVNYTQLAGMHASYADKGLRILGFPCNQFGGQEPGTEVEIKEFVKQFDVQFDMFSKIDVNGDGAHPLFKWMKEQPKG

Salmon GPx4 54 GKTRVNYTQLEGMHASYADKGLRILGFPCNQFGGQEPGTEVEIKEFVKQFDVQFDMFSKIDVNGDGAHPLFKWMKEQPKG

Zebraf ishGPx4b 68 GKTPVNYTQLAAMHVTYAEKGLRILGFPCNQFGKQEPGSEAEIKEFAKGYNAEFDLFSKIDVNGDAAHPLWKWMKEQPKG

MouseG Px4 74 GKTDVNYTQLVDLHARYAECGLRILAFPCNQFGRQEPGSNQEIKEFAAGYNVKFDMYSKICVNGDDAHPLWKWMKVQPKG

HumanG Px4 74 GKTEVNYTQLVDLHARYAECGLRILAFPCNQFGKQEPGSNEEIKEFAAGYNVKFDMFSKICVNGDDAHPLWKWMKIQPKG

TroutG Px4a1 157 KGFLGNGIKWNFTKFLINREGQVMKRYGPMDDPSVVEKDLPKYL--

TroutG Px4a2 157 RSFLGNGIKWNFTKFLINREGQVMKRYGPMGDPSAVEKDLPKYL--

Zebraf ishGPx4a 143 KGFLGNGIKWNFTKFLINREGQVVKRYSPLQDPSVVEKDLSKYL--

TroutG Px4b 148 KGTLGNNIKWNFTKFLINREGQVVKRYGPMDDPIVIEKDLPKYL--

Salmon GPx4 134 KGTLGNNIKWNFTKFLINREGQVVKRYGPMDDPIVIEKDLPKYLSL

Zebraf ishGPx4b 148 RGTLGNNIKWNFTKFLIDREGQVVKRYGPMDDPSVVEKDLPKYL--

MouseG Px4 154 RGMLGNAIKWNFTKFLIDKNGCVVKRYGPMEEPQVIEKDLPCYL--

HumanG Px4 154 KGILGNAIKWNFTKFLIDKNGCVVKRYGPMEEPLVIEKDLPHYF--

Fig. 3. Multiple alignment of the deduced amino acid sequences of trout GPx1 isoforms with known ﬁsh GPx1 molecules and selected mammalian GPx1 and GPx2 molecules

(A), and trout GPx4 isoforms with selected GPx4 molecules from ﬁsh, and mammals (B). The multiple alignment was produced using ClustalW2 and conserved amino acids

shaded using BOXSHADE (V3.21). The accession numbers for sequences used in this analysis are given in Fig. 1. Amino acid residues of the catalytic triad are marked with

an asterisk, with the seleocysteine (SeCys) residue (U) marked with an arrow. Please, note that the salmon GPx4 is from an EST with an incompletely sequenced N-terminal

region, and a possible error at the C-terminus.

Interestingly, comparing the transcriptional induction of GPx1 relative to the inorganic form) all the GPx1 isoforms were still

isoforms within the Cys-Se-Se-Cys treatments with the induction more highly induced by the organic Se. For example, comparing

of the same genes in cells treated with ten times higher concentra- GPx1a at 100 nM Cys-Se-Se-Cys versus 1 ␮M Na2SeO3 and at 1 ␮M

tions of Na2SeO3 (since the organic form has a higher Se content Cys-Se-Se-Cys versus 10 ␮M Na2SeO3 it is clear that the organic

D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111 105

GPx 1a

A. GPx1

GPx1b 1

1,000 ,000

GPx1b 2

100,00 0

10,00 0 1,00 0

Relative expression 100

1 Skin Liver Gills Brain Heart Blood Gonad Scales Spleen Thymus Tail fins Tail adip Tail Stomach

Esophagus

Mid intestine Head Kidney Pyloric caeca Pyloric Distal intestine Distal

GPx1a - GPx1b1 .027 .300.002.424 .123.398.238.477.017.653.168.278 .148.000.000.002 .000 .006

GPx1a - GPx1b2 .019 .304.001.219 .000.637.938.209.020.001.166.158 .001.001.253.001 .036 .001

GPx1b1 - GPx1b2 .157 .657.906.504 .625.234.846.204.100.007.684.195 .045.000.003.000 .000 .002

Ratio GPx1-GPx4 0.12 0.110.130.120.58 0.320.081.711.02 1.64 0.190.530.310.01 0.020.040.45 0.26

GPx4a 1

B. GPx4 GPx4a 2

1,000 ,000

GPx4b

100,00 0

10,00 0 1,00 0

100 Relative expression

1 Skin Liver Gills Brain Heart Bloo d Scales Gonad Spleen Tail fins Thymus Tail adip Tail Stomach

Esophagus

Mid intestine Head Kid ney Pyloric caeca Distal intestine Distal

GPx4a1 - GPx4a2 .002 .000.000 .001.000 .000.008.000.000.000.000.002.000 .000 .001 .001.001 .001

GPx4a1 - GPx4b .016 .492.027 .407.020 .364.157.863.893.863.052.4470.28 .023 .017 .003.012 .009

GPx4a2 - GPx4b .029 .008.007 .001.002 .004.017.001.001.001.000.001.000 .001 .001 .000.002 .001

Ratio GPx4-GPx1 3.8 11 7.4 6.6 1.63.1 12 0.40.90.65.21.73.3 143 42 22 2.23.8

Fig. 4. Relative transcript expression of trout GPx1 and GPx4 isoforms in vivo. The transcript expression level of GPx1a, GPx1b1 and GPx1b2 (A), and GPx4a1, GPx4a2 and GPx4b

(B) was determined by real time PCR in 18 tissues from six ﬁsh and expressed as arbitrary units that were ﬁrst normalized against the transcript expression level of EF-1˛.

For comparison, the average transcript expression level of GPx4b1 in liver, which has the lowest level among the six GPx genes in all the tissues examined, was deﬁned as 1.

The results were presented as the mean + SEM. The p value of a paired-sample t test comparing the expression of different isoforms of GPx1 or GPx4 is shown below, where

signiﬁcant differences between pairs of genes are shaded. The ratio of the average expression of the GPx1 and GPx4 isoforms in each tissue is also shown below.

form is more effective. Similarly, GPx1b1 was more highly induced treated with 10 ␮M Na2SeO3. Among the GPx1 isoforms, GPx1b1

at 10 nM Cys-Se-Se-Cys versus 100 nM Na2SeO3, at 1 ␮M Cys-Se- was the only form not responsive to inorganic selenium at the lower

␮

Se-Cys versus 10 M Na2SeO3 and at 10 ␮M Cys-Se-Se-Cys versus concentrations of 10 nM and 100 nM.

50 ␮M Na2SeO3. Transcript expression of GPx1b2 gene was the The GPx4 isoforms appeared to be less sensitive to Se expo-

most sensitive to organic Se, where the level of induction detected sure. Amongst these GPx4a1 was the most sensitive, in that it

at 10 nM Cys-Se-Se-Cys was equivalent to that reached in the cells was the only isoform signiﬁcantly induced by Se at the lowest

106 D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111

A. GPx1a D. GPx4a1 5.0 a 5.0

4.5 4.5

4.0 ab 4.0 3.5 bc 3.5

c 3.0 c c c 3.0 c c 2.5 2.5

Fold change a 2.0 Fold change Fold change 2.0 a a a a d a a 1.5 1.5

1.0 1.0

0.5 0.5

0.0 0.0

CTRL 10nM 100nM 1uM 10uM 50uM CTRL 10nM 100nM M10M50M Concentration of selenocompound Concentration of selenocompound

B. GPx1b1 E. GPx4a2 5.0 5.0

4.5 4.5

4.0 a 4.0

3.5 3.5

3.0 3.0 a a a a a 2.5 2.5 a b a a bc bc c 2.0 2.0 Fold change cd ce Fold change 1.5 f de 1.5 f 1.0 1.0

0.5 0.5

0.0 0.0

CTRL 10nM 100nM 1uM 10uM 50uM CTRL 10nM 100nM M10M50M Concentration of selenocompound Concentration of selenocompound

C. GPx1b2 F. GPx4b 15.0 5.0 14.0 4.5 13.0 a 12.0 4.0 11.0 3.5 10.0 9.0 3.0 8.0 b 2.5 7.0

Fold change 6.0

Fold change 2.0 5.0 c 1.5 4.0 d 3.0 d d d e 1.0 2.0 e e 0.5 1.0

0.0 0.0

CTRL 10nM 100nM M10M50M CTRL 10nM 100nM M10M50M

Concentration of selenocompound Concentration of selenocompound

Fig. 5. Transcriptional modulation of trout GPx1a (A), GPx1b1 (B), GPx1b2 (C), GPx4a1 (D), GPx4a2 (E) and GPx4b (F) in the RTL cell line stimulated with Na2SeO3 and Cys-Se-Se-

Cys at different concentrations, or medium as control. The treatments were terminated at 24 h post-stimulation and total RNA was isolated. The transcript expression of the

GPx genes in RTL cells was quantiﬁed by real time PCR and normalised to the transcript expression level of EF-1˛ from the same sample, and then used for statistical analysis.

A fold change, calculated as the average expression level of stimulated samples divided by that of the controls, is presented. The results represent the mean + SEM from

four independent cultures. Values are statistically signiﬁcant from the controls (p < 0.05) where marked with a letter, with different letters indicating signiﬁcant differences

between the treatment groups.

concentration (10 nM) used of the organic form. However, its induc- enzymatic activity detectable in the cell lysates. Although the

␮

tion returned to control levels at the highest concentration (50 M) three GPx1 genes showed different patterns of responsiveness to

␮

tested. At concentrations of 100 nM, 1 M and 10 ␮M both GPx4a1 exposure to the two selencompounds studied, the increases in

and GPx4a2 were similarly induced by the inorganic and organic Se transcript level did not coincide with any signiﬁcant increase in

treatments, with maximal increases of approx. 2-fold. GPx4a2 was GPx1 activity (Supplementary Fig. S2).

the only responsive gene at the highest concentration used. In con-

trast, GPx4b did not show any signiﬁcant difference in expression 3.5. Effect of Na2SeO3 and Cys-Se-Se-Cys on cell viability

level at any concentration tested relative to the control cells.

Since GPx1 genes were the most responsive under the con- Cell viability was evaluated by determining the cell mem-

ditions tested, the GPx1 enzymatic activity was also analyzed, brane integrity, using the LDH assay (Fig. 6). The cell membrane

to see if these increases in transcript level were reﬂected in the integrity was comparable to control values in cells treated with

D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111 107

100

normal feed diet used in salmonid farming provide a range between

−1

90 1.5–3.3 ␮g Se g (dry mass) (Julshamn et al., 1990; Lorentzen et al.,

80 1994). Waterborne Se can also contribute to Se bioaccumulation

in ﬁsh. Usually the concentration of Se found in the water column 70

−1

is relatively low, approximately 0.1–0.4 ␮g Se l , but in polluted

60 −1

conditions concentrations of 40–130 ␮g Se l can exert toxicity

(Watanabe et al., 1997).

40 a A wide range of levels for Se requirement/toxicity in ﬁsh has

30 been reported (Hamilton, 2004). This range is likely due to the

Cytotoxicity (%)

20 various types of selenocompounds assimilated by the ﬁsh. Recent

studies have demonstrated that organic selenocompounds exhibit

several advantages over inorganic forms, such as increased Se

retention, higher resistance to stress (e.g. crowding conditions,

-10

hypoxia and handling), lower induction of lipid peroxidation and

increased antioxidant capacity (KüÜkbay et al., 2009; Rider et al., 1µM 10nM 10µM 50µM

100nM

2009). In addition, different required/tolerated levels are often

Concentration of selenocompound

inﬂuenced by the experimental conditions and which tissues and

biomarkers are used. In ﬁsh the anterior intestine is believed to

Fig. 6. Viability of RTL cells incubated with Na2SeO3 and Cys-Se-Se-Cys at differ-

ent concentrations. The treatments were terminated at 24 h post-exposure and the be the principal route for dietary Se assimilation, and the liver the

cell viability was determined by analysing cell membrane integrity, using the LDH main organ for metabolism. The kidney can also accumulate signif-

assay. Data represent the mean expressed as percentage of the control (0 nM) ± SEM

icant amounts of selenium during the process of excretion (Hodson

of three different cultures. Values are statistically signiﬁcant from the controls

and Hilton, 1983). A recent meta-analysis study reviewed the toxic

(p < 0.05) where marked with a letter, with different letters indicating signiﬁcant

thresholds of Se in the diet and body, and highlighted the impor-

differences between the treatment groups.

tance of considering ovary-based threshold levels (DeForest et al.,

1999). Indeed, the primary effect of selenium on ﬁsh populations

is manifested via maternal transfer from the ovaries to the eggs

both inorganic and organic Se up to a concentration of 1 ␮M. At

(Gillespie and Baumann, 1986; Woock et al., 1987; Schultz and

10 ␮M and over, a marked difference between the two treatments

Hermanutz, 1990; Hermanutz et al., 1992), with the dietary path-

was recorded. Integrity was slightly reduced (by 4.5%) using the

way being the most important exposure route for juvenile and adult

inorganic Se at a concentration of 50 ␮M, but with the organic Se

ﬁsh (Sandholm et al., 1973; Bertram and Brooks, 1986; Besser et al.,

treatment membrane integrity was reduced by 7.5% and 30% at

1993).

concentrations of 10 ␮M and 50 ␮M respectively, although only

The symptoms of Se toxicity in ﬁsh have been extensively

this last result was statistically signiﬁcant from the control.

reviewed (Lemly, 1998, 2002). Most of the studies have determined

the required and toxic range of Se in ﬁsh, based on mortality, growth

4. Discussion abnormalities and histological survey data. Also, the activity of

antioxidant enzymes, such as GPx and other markers of oxidative

It is recognized that selenium is an essential component for sev- stress, have been extensively used in ﬁsh both in vitro and in vitro

eral biological pathways at the molecular, cellular and systemic (Dörr et al., 2008; Misra and Niyogi, 2009; Rider et al., 2009, 2010;

level in every living organism (Neve, 1991). Selenium deﬁciency Elia et al., 2011; Misra et al., 2012a). Nevertheless, more in depth

can heavily compromise the health of an organism, leading to car- investigations on the mechanisms of Se uptake and its metabolism

diovascular pathology, disruption of reproduction, impairment of at the cellular level in ﬁsh are warranted. Although the metabolism

immune defences and as a consequence of this pathogen outbreaks of selenocompounds in zebraﬁsh (Danio rerio) has been described

(Rayman, 2000). Due to the importance of this trace element in bio- in the KEGG (Kyoto Encyclopedia of Genes and Genome) metabolic

logical systems, the use of selenium as a food and feed supplement pathways database, functional characterization at the metabolite

in animal farming, including aquaculture, is increasing (Watanabe level has not been completed (Misra et al., 2010). Without this

et al., 1997; Kiron, 2012). In the aquatic environment, selenium is information on Se metabolism, it is still difﬁcult to be sure of the

widely distributed and a concern regarding elevated levels of Se optimal Se requirement in the diet and to evaluate which form is

for the biota is its ability to bioaccumulate through the food. Thus, the most bioavailable and least toxic. Identiﬁcation and character-

at levels above those required selenium can have adverse effects. ization of genes associated with Se metabolism (i.e. selenoprotein

In animals, Se toxicity produces conditions know as “blind stag- genes) in ﬁsh may allow a better understanding of the impact of

gers” and “alkali disease”, that result in tissue lesions, reproductive Se on homeostasis in these animals. Also, the analysis of Se-related

abnormalities and even death (O‘Toole and Raisbeck, 1995; O‘Toole gene expression can represent a sensitive biomarker to monitor Se

et al., 1996). intake in ﬁsh and potentially redeﬁne the threshold of requirement

Dietary requirements for ﬁsh have been reported to range vs toxicity for different selenocompounds.

−1

between 0.1–0.5 ␮g Se g (dry mass) (Hodson and Hilton, 1983; In this study we have cloned and characterized several rainbow

Gatlin and Wilson, 1984). However, recent investigations had trout, Oncorhynchus mykiss, GPx1 and GPx4 genes. The clustering of

demonstrated that the Se required level may increase up to the sequences in the phylogenetic tree conﬁrmed the association of

−1

4.0 ␮g Se g (dry mass) in ﬁsh subject to stress conditions the sequences characterized in trout with the respective GPx clades.

(KüÜkbay et al., 2009; Rider et al., 2009). Previous studies car- Interestingly three isoforms for both GPx1 (GPx1a, 1b1 and 1b2) and

ried out in trout, reported that this species in normal condition GPx4 (GPx4a1, a2 and b) genes have been identiﬁed in this species.

could regulate Se accumulation through the excretion of up Previously two isoforms of GPx1 (GPx1a and 1b) and GPx4 (GPx4a

␮ −1

to 3.0 g Se g (dry mass), however over this level Se rapidly and b) have been found in teleost ﬁsh, such as Danio rerio (Kryukov

bioaccumulated and caused chronic toxicity to ﬁsh. For example, and Gladyshev, 2000). The constructed phylogenetic tree indicates

−1

elevated mortality has been recorded in ﬁsh fed over 13.0 ␮g Se g the ﬁsh GPx4 is an orthologue of the mammalian molecule, whereas

(Hilton et al., 1980). It has been documented that crustacean prey the ﬁsh GPx1 molecules are not clear orthologues of mammalian

−1

of salmonids may contain concentrations ≥3.0 ␮g Se g and the GPx1 or GPx2. GPx2 (gastro-intestinal glutathione peroxidase,

108 D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111

GI-GPx) is characterized only in mammalian models so far (Chu and GPx1b2 showed a higher up-regulation at increasing con-

et al., 1993) and it is predominantly expressed in proliferative centrations, with maximum transcript expression reached at the

zones of the intestinal mucosa (Florian et al., 2001). If the ﬁsh GPx1 highest concentration used (50 ␮M) for both organic and inorganic

molecule is related to an ancestral molecule that subsequently gave selenium. The transcriptional up-regulation of GPx1 genes was sig-

rise to mammalian GPx1 and GPx2 it is possible that functional niﬁcantly higher with the organic selenium treatments at most of

divergence of the different ﬁsh paralogs may have occurred in an the concentrations tested, even when comparing the induction by

analogous manner. In this context, our results from the tissue dis- Cys-Se-Se-Cys treatment to cells exposed to a ten times higher

tribution analysis of the trout GPx1 isoforms perhaps suggest that concentration of Na2SeO3. The organic form had more effects on

the trout GPx1b1 isoform may be more related functionally to GPx2. transcript expression of GPx1 isoform and GPx1b2 was clearly the

It is recognized that teleost ﬁsh possess a large selenoproteome most induced by organic selenium exposure.

(Lobanov et al., 2009) and our results hint that salmonids may have The viability assay showed a reduction of membrane integrity

the biggest set of selenoproteins amongst all vertebrates. The pres- at the highest concentration of Cys-Se-Se-Cys. The present results

ence of two copies of GPx1 (GPx1a and GPx1b) and GPx4 (GPx4a and suggest that Cys-Se-Se-Cys can be toxic to ﬁsh cells in vitro above a

GPx4b) genes in zebraﬁsh and other teleosts is likely the result of a concentration of 10 ␮M, whereas Na2SeO3 was only found to have

third-round whole genome duplication (WGD) event that occurred a toxic effect (albeit small) at 50 ␮M. The common assumption in

during teleost evolution (Dehal and Boore, 2005; Volff, 2005). After- selenium cytoxicity studies is that inorganic selenocompounds are

wards, a fourth round of WGD occurred in Salmonid ﬁsh, which less bioavailable and more toxic at higher concentrations, whereas

underwent a diploidization of their chromosomic set. As pseudo- organic forms are more bioavailable and tolerated (Hoeﬁg et al.,

tetraploid organisms, Salmonids (e.g. rainbow trout and Atlantic 2011). However, it is also possible that Cys-Se-Se-Cys being more

salmon) may possess three to four copies of genes present as a sin- bioavailable can reach higher concentrations within a cell, which

gle copy in mammals and in duplicate in other bony ﬁsh (Ohno ultimately induce selenium toxic effects. The presence of two sele-

et al., 1968; Allendorf and Thorgaard, 1984). Sequence analysis nium residues also can contribute to its adverse effects on cell

conﬁrmed the presence of the catalytic triad, including the seleno- viability. Previous studies in vivo have shown that Cys-Se-Se-Cys is

cysteine residue, within the trout GPx1 and GPx4 proteins, which able to induce toxicity in mice hepatocytes, by inducing the forma-

are universally conserved in all phyla (Herbette et al., 2007). The tion of ROS and inhibiting the suppression of the Se methylation

presence of a SECIS element within the 3 -UTR was also conﬁrmed (the main mechanism to metabolise and detoxify selenocom-

in all the sequences identiﬁed. In the GPx1 isoforms, the SECIS ele- pounds in cells) (Hasegawa et al., 1996). Another study conducted

ment was formed by the motif 5 TGAA3 :5 TGAT3 , whilst in GPx4 in vitro showed that Cys-Se-Se-Cys had cytotoxic effects on splenic

5 TGAC3 :5 TGAT3 was used. lymphocytes isolated from mice in the concentration range 250 nM

As GPx1 and GPx4 are important components of the intracel- to 100 ␮M (Santhosh Kumar et al., 2009). To date few studies have

lular antioxidant defences of many cell types, we analysed where been conducted to determine the therapeutic properties versus the

these genes are tanscribed in a wide range of tissues. GPx4 iso- toxic effects of Cys-Se-Se-Cys, especially in ﬁsh models (Bjerregaard

forms were highly transcribed in all tissues examined compared et al., 2011; Huang et al., 2012a; Huang et al., 2012b). Further stud-

with GPx1 isoforms, with the exception of blood, head kidney ies are clearly required to better elucidate the metabolism and

and spleen. The higher constitutive transcription of GPx1 in the biological effects of this diselenocompound, both in vitro and in

two main hematopoietic tissues in ﬁsh (head kidney and spleen) vitro.

(Catton, 1951; Ellis et al., 1978) may be associated to the high A comparison of the results obtained from the expression anal-

amount of ROS (especially H2O2) produced during hematopoiesis ysis and the viability assay suggests that the GPx1a up-regulation

and their essential role in the regulation of hematopoietic growth detected at concentrations ranging from 10 nM to 1 ␮M is only

factor (HGF) signalling (Sattler et al., 1999; Ghaffari, 2008). The related to the selenium availability, whereas the notable up-

high level of GPx1 in the blood may be associated with the high regulation of GPx1b1 and GPx1b2 at the highest concentrations

oxidative stress that occurs in erythrocytes, being continuously used (10 and 50 ␮M) may be a direct effect of cell stress caused

exposed to ROS as the main oxygen carriers in the organism (Mills, by selenium. Since trout have a larger number of GPx1 genes,

1957). The transcript expression of GPx4 (GPx4a1) was particularly different roles of the three isoforms is likely through subfunctional-

high in the gut, a site that is likely to be continuously exposed isation, with GPx1b1 and GPx1b2 apparently induced preferentially

to pro-oxidant compounds and oxidative species in water com- in response to oxidative stress.

ing from the external environment, with the potential to damage The higher transcript response to Se of trout GPx1 compared to

cell membranes (Wilhelm et al., 1993; Wilhelm, 2007). In addition, the GPx4 genes supports the hypothesis that within the selenopro-

the stomach and pyloric cecae are involved in the digestion and teome and even within the same selenoprotein family there are

absorption of nutrients, which may also easily contain exogenous proteins with an essential role in cellular function but that are less

compounds with oxidative properties, such as iron, that can react responsive to selenium availability, as well as proteins that are non

with ascorbate released during digestion to give rise to ROS for- essential but are more involved in protection from environmental

mation through Fenton chemistry (Kadiiska et al., 1995; Halliwell stress and selenium availability (Carlson et al., 2005). Indeed,

and Gutteridge, 1999). Whilst trout GPx1 and GPx4 isoforms were gene knock out studies in mice have shown that GPx4 (Yant et al.,

expressed in every tissue examined, GPx4a2 showed the lowest 2003), along with other selenoproteins such as TrxR3 and DIO2

transcript expression, with the lowest transcript expression level (Schneider et al., 2001; Conrad et al., 2004), has an essential role

in the liver. in cellular function and removal is lethal or results in an abnormal

The effect of Se availability on the transcript expression of phenotype. In contrast, mammalian GPx1 (Cheng et al., 1998) and

GPx1 and GPx4 genes was tested in vitro. Preliminary experi- GPx2 (Esworthy et al., 2001) probably have non-essential func-

ments were conducted using a wide range of concentrations (from tions since their removal in knockout mice manifests little or no

␮

1 nM to 200 M, in 1:10 serial dilution) to define an optimal phenotypic modification. In selenium deficient rats GPx1 activity

∼

range where effects on gene transcription following Se exposure is reduced to 1% of normal levels in liver, and to 4 ± 9% in kidney,

could be detected but severe adverse effect on cell viability were heart and lung, whilst GPx4 activity is decreased to 25 50%

avoided. The GPx1 isoforms showed a relevant induction in all in the same tissues and is unaffected by selenium deprivation

the selenium treatments. GPx1a was more induced by selenium in testes. The dramatic decline in GPx1 activity upon selenium

at low/intermediate concentrations up to 1 ␮M, whereas GPx1b1 deprivation is due in large part to a rapid turnover of the mRNA for

D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111 109

this protein, and transcriptional control of expression (Saedi et al., effect on GPx1 enzymatic activity was detected. Thus, we conclude

1988; Christensen and Burgener, 1992; Xin et al., 1995). that transcript expression of trout GPx1 gene may be a sensitive

Although the transcript expression of GPx1 isoform was signif- biomarker of selenium exposure in vitro, and may aid the evaluation

icantly modulated in this study by the two selenocompounds, no of the impact of selenium concentration and chemical speciation on

signiﬁcant change was seen on the enzymatic activity of the two dif- cell homeostasis. In addition, the in vitro approach used here has

ferent treatments at the concentrations tested. The absence of any allowed us to describe, characterize and delimit Se effects, and gives

induction of GPx1 activity upon exposure to Na2SeO3 at concen- complementary information for further investigations in vivo.

trations up to 50 ␮M conﬁrmed previous results obtained in trout

isolated hepatocytes exposed to different concentrations (ranging Acknowledgements

from 50 to 200 ␮M) of Na2SeO3 for 24 h (Misra and Niyogi, 2009).

However, studies of the effect of Cys-Se-Se-Cys on GPx1 activity This work was supported by Alltech and the Marine Alliance

in vitro are lacking. A study conducted on trout isolated hepatocytes for Science and Technology Scotland (MASTS). We also thank Dr.

exposed to different concentrations of selenomethionine (SeMet) Andrea Raab (TESLA) for providing support with the ICP-MS anal-

for 24 h showed that the GPx1 activity was slightly induced only ysis.

at a concentration equal to 1000 ␮M (Misra et al., 2012a). How-

ever, it is not possible to compare SeMet and Cys-Se-Se-Cys because

Appendix A. Supplementary data

ﬁrstly they are a mono- and di-selenocompound, respectively, and

secondly (as discussed above) every selenocompound may have

Supplementary data associated with this article can be

different bioavailability and bioactivity.

found, in the online version, at http://dx.doi.org/10.1016/

The reason why there was no change in GPx1 activity may j.aquatox.2012.12.020.

relate to it being a homotetramer, with even higher fold-increases

in transcript level being needed to inﬂuence the activity of the References

enzyme. Moreover, GPx1 contributes only partially to the global

enzymatic activity and this could explain the differences observed Allendorf, F.W., Thorgaard, G.H., 1984. Tetraploidy and the evolution of salmonid

ﬁshes. Evolutionary Genetics of Fishes, 1–53.

between mRNA level and enzymatic activity. Furthermore, a differ-

Altschul, S.F., Madden, T.L., Schäffer, A.A., Zhang, J., Zhang, Z., Miller, W., Lipman,

ent sensitivity towards the inducers at the transcript level, mRNA

D.J., 1997. Gapped BLAST and PSI-BLAST A new generation of protein database

processing, or transport and protein stability may lead to discrep- search programs. Nucleic Acids Research 25, 3389–3402.

ancies between transcript expression and enzyme activity (Okey, Anundi, I., Stahl, A., Hogberg, J., 1984. Effects of selenite on O2 consumption glu-

tathione oxidation and NADPH levels in isolated hepatocytes and the role of

1990). Finally, in this particular study the different GPx1 isoforms

redox changes in selenite toxicity. Chemico-Biological Interactions 50, 277–288.

showed two different kinds of responses, one related to the Se

Apel, K., Hirt, H., 2004. Reactive oxygen species Metabolism, oxidative stress, and

availability and the other to Se-induced stress. These two differ- signal transduction. Annual Review of Plant Biology 55, 373–399.

Arduser, F., Wolffram, S., Scharrer, E., 1985. Active absorption of selenate by rat

ent patterns of transcript expression make it difﬁcult to deﬁne a

ileum. Journal of Nutrition 115, 1203–1208.

trend at the level of enzymatic activity. However, our results are

Arthur, J.R., 2000. The glutathione peroxidases. Cellular and Molecular Life Sciences

consistent with Ideker et al. (2001) and Grifﬁn et al. (2002) who 57, 1825.

2 Beilstein, M.A., Whanger, P.D., 1992. Selenium metabolism and glutathione peroxi-

found a low correlation coefﬁcient (r = 0.61) between mRNA and

dase activity in cultured human lymphoblasts Effects of transsulfuration defects

protein level changes depending on the gene investigated. Tukey

and pyridoxal phosphate. Biological Trace Element Research 35, 105–118.

and Johnson (1990) also postulated a different turnover and dura- Berry, M.J., Banu, L., Chen, Y., Mandel, S.J., Kiefer, J.D., Harney, J.W., Larsen, P.R., 1991.

Recognition of UGA as a selenocysteine codon in Type I deiodinase requires

tion of induction between gene expression and enzyme activity.

sequences in the 3’ untranslated region. Nature 353, 273–276.

To our knowledge, only one previous study has investigated the

Berry, M.J., Banu, L., Harney, J.W., Larsen, P.R., 1993. Functional characterization of

correlation between GPx1 transcript level and enzymatic activity the eukaryotic SECIS elements which direct selenocysteine insertion at UGA

codons. EMBO Journal 12, 3315–3322.

in a ﬁsh model, and in agreement with our results they did not

Bertram, P.E., Brooks, A.S., 1986. Kinetics of accumulation of selenium from food and

detect any change at the activity level despite a signiﬁcant mod-

water by fathead minnows. Water Research 20, 877–884.

ulation of transcription (Doyen et al., 2012). Thus, further studies Besser, J.M., Canﬁeld, T.J., La Point, T.W., 1993. Bioaccumulation of organic and

are required to quantify at what point the level of gene transcrip- inorganic selenium in a laboratory food chain. Environmental Toxicology and

Chemistry 12, 57–72.

tion inﬂuences the activity of the protein. Also, at this stage it is not

Birringer, M., Pilawa, S., Flohé, L., 2002. Trends in selenium biochemistry. Natural

possible to determine if the different isoforms contribute equally,

Product Reports 19, 693–718.

or disproportionately, to the synthesis and activity of the protein. Bjerregaard, P., Fjordside, S., Hansen, M.G., Petrova, M.B., 2011. Dietary selenium

reduces retention of methyl mercury in freshwater ﬁsh. Environmental Science

& Technology 45, 9793–9798.

Bock, A., Forchhammer, K., Heider, J., Baron, C., 1991. Selenoprotein synthesis: an

5. Conclusions expansion of the genetic code. Trends in Biochemical Sciences 16, 463–467.

Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of micro-

gram quantities of protein utilizing the principle of protein dye binding.

In summary we have cloned three paralogs of GPx1 (GPx1a,

Analytical Biochemistry 72, 248–254.

GPx1b1 and GPx1b2) and GPx4 (GPx4a1, GPx4a2 and GPx4b) in Burk, R.F., Hill, K.E., 1993. Regulation of selenoproteins. Annual Review of Nutrition

13, 65–81.

rainbow trout. The sequences have been analyzed with bioinfor-

Campanella, J., 2003. MatGAT: an application that generates similarity/identity

matics tools to conﬁrm their identiﬁcation as genes encoding GPx

matrices using protein or DNA sequences. BMC Bioinformatics 4, 29.

proteins; by homology and phylogenetic tree analysis, the conser- Carlson, B.A., Xu, X., Gladyshev, V.N., Dolph, L.H., 2005. Selective rescue of seleno-

protein expression in mice lacking a highly specialized methyl group in

vation of the catalytic sites and the presence of a SECIS element

selenocysteine tRNA. Journal of Biological Chemistry 280, 5542–5548.

in the 3 -UTR, fundamental characteristics of every selenoprotein.

Catton, W.T., 1951. Blood cell formation in certain teleost ﬁshes. Blood 6, 39–60.

Constitutive transcript expression was found in 18 different tis- Chaudiere, J., Courtin, O., Leclaire, J., 1992. Glutathione oxidase activity of seleno-

sues studied, with GPx4 more highly transcribed in most tissues cystamine A mechanistic study. Archives of Biochemistry and Biophysics 296,

328–336.

examined (except blood, head kidney and spleen). With the excep-

Cheng, W., Ho, Y., Valentine, B.A., Ross, D.A., Combs Jr., G.F., Lei, X.G., 1998. Cellu-

tion of GPx4b, the GPx genes, and especially GPx1 isoforms, were

lar glutathione peroxidase is the mediator of body selenium to protect against

responsive to selenium exposure. GPx1a was the most sensitive to paraquat lethality in transgenic mice. Journal of Nutrition 128, 1070–1076.

Chenna, R., 2003. Multiple sequence alignment with the Clustal series of programs.

selenium availability in non stressful conditions, whereas GPx1b1

Nucleic Acids Research 31, 3497–3500.

and GPx1b2 were highly induced by exposure to selenium levels

Christensen, M.J., Burgener, K.W., 1992. Dietary selenium stabilizes glutathione per-

that had some toxic effects on the cells. However, no signiﬁcant oxidase mRNA in rat liver. Journal of Nutrition 122, 1620–1626.

110 D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111

Chu, F., Doroshow, J.H., Esworthy, R.S., 1993. Expression, characterization, and tis- autoregulatory mechanism in selenocysteine metabolism? Proceedings of the

sue distribution of a new cellular selenium-dependent glutathione peroxidase, National Academy of Sciences of the United States of America 93, 15086–15091.

GSHPx-GI. Journal of Biological Chemistry 268, 2571–2576. Halliwell, B.,Gutteridge, J.M.C., 1999. Free Radicals in Biology and Medicine, Third

Combs Jr., G.F., Combs, S.B., 1986. The Role of Selenium in Nutrition. Academic Press ed. Oxford University Press.

Inc, New York. Hamilton, S.J., 2004. Review of selenium toxicity in the aquatic food chain. Science

Conrad, M., Jakupoglu, C., Moreno, S.G., Lippl, S., Banjac, A., Schneider, M., Beck, of the Total Environment 326, 1–31.

H., Hatzopoulos, A.K., Just, U., Sinowatz, F., Schmahl, W., Chien, K.R., Wurst, Hasegawa, T., Mihara, M., Nakamuro, K., Sayato, Y., 1996. Mechanisms of selenium

W., Bornkamm, G.W., Brielmeier, M., 2004. Essential role for mitochondrial methylation and toxicity in mice treated with selenocystine. Archives of Toxi-

thioredoxin reductase in hematopoiesis heart development, and heart function. cology 71, 31–38.

Molecular and Cellular Biology 24, 9414–9423. Hatﬁeld, D.L., 2002. How selenium has altered our understanding of the genetic code.

Contempre, B., Le Moine, O., Dumont, J.E., Denef, J., Many, M.C., 1996. Selenium deﬁ- Molecular and Cellular Biology 22, 3565–3576.

ciency and thyroid ﬁbrosis A key role for macrophages and transforming growth Hefnawy, A.E.G., Tórtora-Pérez, J.L., 2010. The importance of selenium and the effects

␤

factor (TGF-␤). Molecular and Cellular Endocrinology 124, 7–15. of its deﬁciency in animal health. Small Ruminant Research 89, 185–192.

Copeland, P.R., Fletcher, J.E., Carlson, B.A., Hatﬁeld, D.L., Driscoll, D.M., 2000. A novel Herbette, S., Roeckel-Drevet, P., Drevet, J.R., 2007. Seleno-independent glutathione

RNA binding protein SBP2, is required for the translation of mammalian seleno- peroxidases: More than simple antioxidant scavengers. FEBS Journal 274,

protein mRNAs. EMBO Journal 19, 306–314. 2163–2180.

Dallinger, R., Prosi, F., Segner, H., Back, H., 1987. Contaminated food and uptake of Hermanutz, R.O., Allen, K.N., Roush, T.H., Hedtke, S.F., 1992. Effects of elevated sele-

heavy metals by ﬁsh: a review and a proposal for further research. Oecologia 73, nium concentrations on bluegills (Lepomis macrochirus) in outdoor experimental

91–98. streams. Environmental Toxicology and Chemistry 11, 217–224.

DeForest, D.K., Brix, K.V., Adams, W.J., 1999. Critical review of proposed residue- Hilton, J.W., Hodson, P.V., Slinger, S.J., 1980. The requirement and toxicity of sele-

based selenium toxicity thresholds for freshwater ﬁsh. Human and Ecological nium in rainbow trout (Salmo gairdneri). Journal of Nutrition 110, 2527–2535.

Risk Assessment 5, 1187–1228. Hodson, P.V., Hilton, J.W., 1983. The Nutritional Requirements and Toxicity to Fish

Dehal, P., Boore, J.L., 2005. Two rounds of whole genome duplication in the ancestral of Dietary and Waterborne Selenium. Ecological Bulletins 35, 335–340.

vertebrate. PLoS Biology 3, e314. Hoeﬁg, C.S., Renko, K., Köhrle, J., Birringer, M., Schomburg, L., 2011. Comparison of

Dörr, A.J.M., Pacini, N., Abete, M.C., Prearo, M., Elia, A.C., 2008. Effects of a selenium- different selenocompounds with respect to nutritional value vs. toxicity using

enriched diet on antioxidant response in adult crayﬁsh (Procambarus clarkii). liver cells in culture. Journal of NutritionalBiochemistry 22, 945–955.

Chemosphere 73, 1090–1095. Huang, S.S.Y., Strathe, A.B., Hung, S.S.O., Boston, R.C., Fadel, J.G., 2012a. Seleno-

Doyen, P., Rodius, F., L‘Hoste, L., Hoffmann, L., Cauchie, H., 2012. Identiﬁcation and compounds in juvenile white sturgeon: Estimating absorption disposition, and

mRNA expression of pi-class glutathione S-transferase and selenium-dependent elimination of selenium using Bayesian hierarchical modeling. Aquatic Toxico-

glutathione peroxidase in the gudgeon Gobio gobio exposed to PCB 77. Compar- logy 109, 150–157.

ative Biochemistry and Physiology C 155, 300–306. Huang, S.S.Y., Strathe, A.B., Wang, W., Deng, D., Fadel, J.G., Hung, S.S.O., 2012b.

Dumont, E., Vanhaecke, F., Cornelis, R., 2006. Selenium speciation from food source Selenocompounds in juvenile white sturgeon: Evaluating blood tissue, and

to metabolites A critical review. Analytical and Bioanalytical Chemistry 385, urine selenium concentrations after a single oral dose. Aquatic Toxicology 109,

1304–1323. 158–165.

Elia, A.C., Prearo, M., Pacini, N., Dörr, A.J.M., Abete, M.C., 2011. Effects of selenium Huber, R.E., Criddle, R.S., 1967. Comparison of the chemical properties of selenocys-

diets on growth accumulation and antioxidant response in juvenile carp. Eco- teine and selenocystine with their sulfur analogs. Archives of Biochemistry and

toxicology and Environmental Safety 74, 166–173. Biophysics 122, 164–173.

Ellis, A.E., Roberts, R.J., Tytler, P., 1978. The anatomy and physiology of teleosts. Fish Ideker, T., Thorsson, V., Ranish, J.A., Christmas, R., Buhler, J., Eng, J.K., Bumgarner,

Pathology, 13–54. R., Goodlett, D.R., Aebersold, R., Hood, L., 2001. Integrated genomic and pro-

El-Sayed, W.M., Aboul-Fadl, T., Lamb, J.G., Roberts, J.C., Franklin, M.R., 2006. Effect of teomic analyses of a systematically perturbed metabolic network. Science 292,

selenium-containing compounds on hepatic chemoprotective enzymes in mice. 929–934.

Toxicology 220, 179–188. Ip, C., Hayes, C., Budnick, R.M., Ganther, H.E., 1991. Chemical form of selenium critical

Esworthy, R.S., Aranda, R., Martín, M.G., Doroshow, G.H., Binder, S.W., Chu, F.F., metabolites, and cancer prevention. Cancer Research 51, 595–600.

2001. Mice with combined disruption of Gpx1 and Gpx2 genes have colitis. Jackson, M.J., 1997. The assessment of bioavailability of micronutrients: introduc-

American Journal of Physiology – Gastrointestinal and Liver Physiology 281, tion. European Journal of Clinical Nutrition 51 (Suppl. 1), S1L S2.

G848–G855. Janz, D.M., 2011. 7–Selenium. In Fish Physiology, Volume 31, Part A. Academic Press,

Fagegaltier, D., Hubert, N., Yamada, K., Mizutani, T., Carbon, P., Krol, A., 2000. Char- pp. 327–374.

acterization of mSelB a novel mammalian elongation factor for selenoprotein Jonnalagadda, S.B., Prasada Rao, P.V.V., 1993. Toxicity, bioavailability and metal spe-

translation. EMBO Journal 19, 4796–4805. ciation. Comparative Biochemistry and Physiology B 106, 585–595.

Finley, J.W., 2006. Bioavailability of selenium from foods. Nutrition Reviews 64, Julshamn, K., Sandnes, K., Lie, O., Waagbo, R., 1990. Effects of dietary selenium

146–151. supplementation on growth, blood chemistry and trace element levels in serum

Fletcher, J.E., Copeland, P.R., Driscoll, D.M., Krol, A., 2001. The selenocysteine incor- and liver of adult Atlantic salmon (Salmo salar). Fiskeridirektoratets skrifter.

poration machinery: Interactions between the SECIS RNA and the SECIS-binding Serie havundersøkelser 3, 47–58.

protein SBP2. RNA 7, 1442–1453. Kadiiska, M.B., Burkitt, M.J., Xiang, Q., Mason, R.P., 1995. Iron supplementation gen-

Flohé, L., Brigelius-Flohé, R., 2012. Selenoproteins of the Glutathione Peroxidase erates hydroxyl radical in vivo An ESR spin- trapping investigation. Journal of

Family, third ed. Springer New York. Clinical Investigation 96, 1653–1657.

Flohé, L., Gunzler, W.A., Schock, H.H., 1973. Glutathione peroxidise: A selenoenzyme. Kiron, V., 2012. Fish immune system and its nutritional modulation for preventive

FEBS Letters 32, 132–134. health care. Animal Feed Science and Technology 173, 111–133.

Florian, S., Wingler, K., Schmehl, K., Jacobasch, G., Kreuzer, O.J., Meyerhof, W., Korzeniewski, C., Callewaert, D.M., 1983. An enzyme-release assay for natural cyto-

Brigelius-Flohé, R., 2001. Cellular and subcellular localization of gastrointestinal toxicity. Journal of Immunological Methods 64, 313–320.

glutathione peroxidase in normal and malignant human intestinal tissue. Free Kryukov, G.V., Castellano, S., Novoselov, S.V., Lobanov, A.V., Zehtab, O., Guigó, R.,

Radical Research 35, 655–663. Gladyshev, V.N., 2003. Characterization of mammalian selenoproteomes. Sci-

Foster, L.H., Sumar, S., 1997. Selenium in health and disease: a review. Critical ence 300, 1439–1443.

Reviews in Food Science and Nutrition 37, 211–228. Kryukov, G.V., Gladyshev, V.N., 2000. Selenium metabolism in zebraﬁsh: Multiplicity

Ganther, H.E., 1986. Pathways of selenium metabolism including respiratory excre- of selenoprotein genes and expression of a protein containing 17 selenocysteine

tory products. Journal of the American College of Toxicology 5, 1–5. residues. Genes to Cells 5, 1049–1060.

Ganther, H.E., 1968. Selenotrisulﬁdes Formation by the reaction of thiols with sele- KüÜkbay, F.Z., Yazlak, H., Karaca, I., Sahin, N., Tuzcu, M., Cakmak, M.N., Sahin, K.,

nious acid. Biochemistry 7, 2898–2905. 2009. The effects of dietary organic or inorganic selenium in rainbow trout

Ganthert, H.E., 1966. Enzymic synthesis of dimethyl selenide from sodium selenite (Oncorhynchus mykiss) under crowding conditions. Aquaculture Nutrition 15,

in mouse liver extracts. Biochemistry 5, 1089–1098. 569–576.

Gasteiger, E., Hoogland, C., Gattiker, A., Duvaud, S., Wilkins, M.R., Appel, R.D., Bairoch, Latrèche, L., 2009. Novel structural determinants in human SECIS elements modulate

A., 2005. Protein Identiﬁcation and Analysis Tools on the ExPASy Server, The the translational recoding of UGA as selenocysteine. Nucleic Acids Research 37,

Proteomics Protocols Handbook, Humana Press, pp. 571–607. 5868–5880.

Gatlin, D.M., Wilson, R.P., 1984. Dietary selenium requirement of ﬁngerling channel Leblondel, G., Mauras, Y., Cailleux, A., Allain, P., 2001. Transport measurements

catﬁsh. Journal of Nutrition 114, 627–633. across Caco-2 monolayers of different organic and inorganic selenium. Biological

Ghaffari, S., 2008. Oxidative stress in the regulation of normal and neoplastic Trace Element Research 83, 191–206.

hematopoiesis. Antioxidants & Redox Signaling 10, 1923–1940. Lee, L.E.J., Clemons, J.H., Bechtel, D.G., Caldwell, S.J., Han, K.B., Pasitschniak-Arts, M.,

Gillespie, R.B., Baumann, P.C., 1986. Effects of high tissue concentrations of selenium Mosser, D.D., Bols, N.C., 1993. Development and characterization of a rainbow

on reproduction by bluegills. Transactions of the American Fisheries Society 115, trout liver cell line expressing cytochrome P450-dependent monooxygenase

208–213. activity. Cell Biology and Toxicology, 9, 279–294.

Grifﬁn, T.J., Gygi, S.P., Ideker, T., Rist, B., Eng, J., Hood, L., Aebersold, R., 2002. Comple- Lemly, A.D., 2002. Symptoms and implications of selenium toxicity in ﬁsh The Belews

mentary proﬁling of gene expression at the transcriptome and proteome levels Lake case example. Aquatic Toxicology 57, 39–49.

in Saccharomyces cerevisiae. Molecular & Cellular Proteomics 1, 323–333. Lemly, A.D., 1998. Pathology of selenium poisoning in ﬁsh. In Environmental Chem-

Guimarães, M.J., Peterson, D., Vicari, A., Cocks, B.G., Copeland, N.G., Gilbert, D.J., Jenk- istry of Selenium. Marcel Dekker, pp. 281–354.

ins, N.A., Ferrick, D.A., Kastelein, R.A., Bazan, J.F., Zlotnik, A., 1996. Identiﬁcation Lobanov, A.V., Hatﬁeld, D.L., Gladyshev, V.N., 2009. Eukaryotic selenoproteins and

of a novel selD homolog from Eukaryotes, Bacteria, and Archaea: Is there an selenoproteomes. Biochimica et Biophysica Acta 1790, 1424–1428.

D. Pacitti et al. / Aquatic Toxicology 130–131 (2013) 97–111 111

Lorentzen, M., Maage, A., Julshamn, K., 1994. Effects of dietary selenite or selenome- results in a phenotype of pituitary resistance to T4. Molecular Endocrinology

thionine on tissue selenium levels of Atlantic salmon (Salmo salar). Aquaculture 15, 2137–2148.

121, 359–367. Schnurr, K., Belkner, J., Ursini, F., Schewe, T., Kühn, H., 1996. The selenoenzyme

Low, S.C., Berry, M.J., 1996. Knowing when not to stop Selenocysteine incorporation phospholipid hydroperoxide glutathione peroxidase controls the activity of the

in eukaryotes. Trends in Biochemical Sciences 21, 203–208. 15-lipoxygenase with complex substrates and preserves the speciﬁcity of the

Margis, R., Dunand, C., Teixeira, F.K., Margis-Pinheiro, M., 2008. Glutathione perox- oxygenation products. Journal of Biological Chemistry 271, 4653–4658.

idase family—An evolutionary overview. FEBS Journal 275, 3959–3970. Schrauzer, G.N., 2000. Selenomethionine: A review of its nutritional signiﬁcance,

Mariotti, M., Ridge, P.G., Zhang, Y., Lobanov, A.V., Pringle, T.H., Guigo, R., Hatﬁeld, metabolism and toxicity. Journal of Nutrition 130, 1653–1656.

D.L., Gladyshev, V.N., 2012. Composition and evolution of the vertebrate and Schultz, R., Hermanutz, R., 1990. Transfer of toxic concentrations of selenium from

mammalian selenoproteomes. PLoS One 7. parent to progeny in the fathead minnow (Pimephales promelas). Bulletin of

McConnell, K.P., Cho, G.J., 1967. Active transport of L-selenomethionine in the intes- Environmental Contamination and Toxicology 45, 568–573.

tine. American Journal of Physiology 213, 150–156. Spallholz, J.E., 1997. Free radical generation by selenium compounds and their

McConnell, K.P., Cho, G.J., 1965. Transmucosal movement of selenium. American prooxidant toxicity. Biomedical and Environmental Sciences 10, 260–270.

Journal of Physiology 208, 1191–1195. Stadtman, T.C., 1996. Selenocysteine. Annual Review of Biochemistry 65, 83–100.

McKenzie, R.C., Arthur, J.R., Beckett, G.J., 2002. Selenium and the regulation of cell Steve Hsieh, H., Ganther, H.E., 1975. Acid-volatile selenium formation catalyzed by

signaling, growth, and survival Molecular and mechanistic aspects. Antioxidants glutathione reductase. Biochemistry 14, 1632–1636.

& Redox Signaling 4, 339–351. Suzuki, K.T., 2005. Metabolomics of selenium: Se metabolites based on speciation

Mills, G.C., 1957. Hemoglobin catabolism. I. Glutathione peroxidase, an erythro- studies. Journal of Health Science 51, 107–114.

cyte enzyme which protects hemoglobin from oxidative breakdown. Journal Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5

of Biological Chemistry 229, 189–197. molecular evolutionary genetics analysis using maximum likelihood, evolu-

Misra, S., Hamilton, C., Niyogi, S., 2012a. Induction of oxidative stress by selenome- tionary distance, and maximum parsimony methods. Molecular Biology and

thionine in isolated hepatocytes of rainbow trout (Oncorhynchus mykiss). Evolution 28, 2731–2739.

Toxicology in Vitro 26, 621–629. Thomas, J.P., Geiger, P.G., Maiorino, M., Ursini, F., Girotti, A.W., 1990. Enzymatic

Misra, S., Kwong, R.W.M., Niyogi, S., 2012b. Transport of selenium across the plasma reduction of phospholipid and cholesterol hydroperoxides in artiﬁcial bilayers

membrane of primary hepatocytes and enterocytes of rainbow trout. Journal of and lipoproteins. Biochimica et Biophysica Acta 1045, 252–260.

Experimental Biology 215, 1491–1501. Toppo, S., Vanin, S., Bosello, V., Tosatto, S.C.E., 2008. Evolutionary and struc-

Misra, S., Niyogi, S., 2009. Selenite causes cytotoxicity in rainbow trout tural insights into the multifaceted glutathione peroxidase (Gpx) superfamily.

(Oncorhynchus mykiss) hepatocytes by inducing oxidative stress. Toxicology in Antioxidants & Redox Signaling 10, 1501.

Vitro 23, 1249–1258. Tukey, R.H., Johnson, E.F., 1990. Molecular aspects of drug metabolizing enzymes. In

Misra, S., Peak, D., Niyogi, S., 2010. Application of XANES spectroscopy in under- Principles of Drug Action, 423–468.

standing the metabolism of selenium in isolated rainbow trout hepatocytes: Ursini, F., Maiorino, M., Valente, M., 1982. Puriﬁcation from pig liver of a protein

Insights into selenium toxicity. Metallomics 2, 710–717. which protects liposomes and biomembranes from peroxidative degradation

Mozier, N.M., McConnell, K.P., Hoffman, J.L., 1988. S-adenosyl-L- and exhibits glutathione peroxidase activity on phosphatidylcholine hydroper-

methionine:thioether S-methyltransferase, a new enzyme in sulfur and oxides. Biochimica et Biophysica Acta 710, 197–211.

selenium metabolism. Journal of Biological Chemistry 263, 4527–4531. Ursini, F., Maiorino, M., Gregolin, C., 1985. The selenoenzyme phospholipid

Nauser, T., Steinmann, D., Koppenol, W.H., 2012. Why do proteins use selenocysteine hydroperoxide glutathione peroxidase. Biochimica et Biophysica Acta (BBA) –

instead of cysteine? Amino Acids 42, 39–44. General Subjects 839, 62–70.

Neve, J., 1991. Physiological and nutritional importance of selenium. Cellular and Valdiglesias, V., Pásaro, E., Méndez, J., Laffon, B., 2010. In vitro evaluation of selenium

Molecular Life Sciences 47, 187–193. genotoxic, cytotoxic, and protective effects. A review. Archives of Toxicology 84,

NRC, 1980. Mineral Tolerance of Domestic Animals. Academic Press. 337–351.

Ohno, S., Wolf, U., Atkin, N.B., 1968. Evolution from ﬁsh to mammals by gene dupli- Volff, J., 2005. Genome evolution and biodiversity in teleost ﬁsh. Heredity 94,

cation. Hereditas 59, 169–187. 280–294.

Okey, A.B., 1990. Enzyme induction in the cytochrome P-450 system. Pharmacology Walczak, R., Westhof, E., Carbon, P., Krol, A., 1996. A novel RNA structural motif in

& Therapeutics 45, 241–298. the selenocysteine insertion element of eukaryotic selenoprotein mRNAs. RNA

O‘Toole, D., Raisbeck, M.F., 1995. Pathology of experimentally induced chronic 2, 367–379.

selenosis (alkali disease) in yearling cattle. Journal of Veterinary Diagnostic Wang, T., Hanington, P.C., Belosevic, M., Secombes, C.J., 2008. Two macrophage

Investigation 7, 364–373. colony-stimulating factor genes exist in ﬁsh that differ in gene organization

O‘Toole, D., Raisbeck, M., Case, J.C., Whitson, T.D., 1996. Selenium-induced blind and are differentially expressed. Journal of Immunology 181, 3310–3322.

staggers and related myths A commentary on the extent of historical livestock Wang, T., Diaz-Rosales, P., Costa, M.M., Campbell, S., Snow, M., Collet, B., Martin,

losses attributed to selenosis on Western US rangelands. Veterinary Pathology S.A., Secombes, C.J., 2011. Functional characterization of a nonmammalian IL-

33, 109–116. 21: rainbow trout Oncorhynchus mykiss IL-21 upregulates the expression of the

Rayman, M.P., 2000. The importance of selenium to human health. Lancet 356, Th cell signature cytokines IFN-gamma, IL-10, and IL-22. Journal of Immunology

233–241. 186, 708–721.

Rider, S.A., Davies, S.J., Jha, A.N., Clough, R., Sweetman, J.W., 2010. Bioavailability of Wang, T., Secombes, C.J., 2003. Complete sequencing and expression of three

co-supplemented organic and inorganic zinc and selenium sources in a white complement components, C1r, C4 and C1 inhibitor, of the classical activation

ﬁshmeal-based rainbow trout (Oncorhynchus mykiss) diet. Journal of Animal pathway of the complement system in rainbow trout Oncorhynchus mykiss.

Physiology and Animal Nutrition 94, 99–110. Immunogenetics 55, 615–628.

Rider, S.A., Davies, S.J., Jha, A.N., Fisher, A.A., Knight, J., Sweetman, J.W., 2009. Watanabe, T., Kiron, V., Satoh, S., 1997. Trace minerals in ﬁsh nutrition. Aquaculture

Supra-nutritional dietary intake of selenite and selenium yeast in normal and 151, 185–207.

stressed rainbow trout (Oncorhynchus mykiss): Implications on selenium status Wendel, A., 1981. Glutathione peroxidase. Methods in Enzymology 77, 325–333.

and health responses. Aquaculture 295, 282–291. Wilhelm, F.D., Giulivi, C., Boveris, A., 1993. Antioxidant defences in marine ﬁsh–I.

Rotruck, J.T., Pope, A.L., Ganther, H.E., Swanson, A.B., Hafeman, D.G., Hoekstra, W.G., Teleosts. Comparative Biochemistry and Physiology C 106, 409–413.

1973. Selenium: Biochemical role as a component of glatathione peroxidase. Wilhelm, F.D., 2007. Reactive oxygen species, antioxidants and ﬁsh mitochondria.

Science 179, 588–590. Frontiers in Bioscience 12, 1229–1237.

Roveri, A., Maiorino, M., Ursini, F., 1994. Enzymatic and immunological mea- Wolffram, S., Anliker, E., Scharrer, E., 1986. Uptake of selenate and selenite by isolated

surements of soluble and membrane-bound phospholipid-hydroperoxide intestinal brush border membrane vesicles from pig, sheep, and rat. Biological

glutathione peroxidase. Methods in Enzymology 233, 202–212. Trace Element Research 10, 293–306.

Saedi, M.S., Smith, C.G., Frampton, J., Chambers, I., Harrison, P.R., Sunde, R.A., 1988. Wolffram, S., Berger, B., Grenacher, B., Scharrer, E., 1989. Transport of selenoamino

Effect of selenium status on mRNA levels for glutathione peroxidase in rat liver. acids and their sulfur analogues across the intestinal brush border membrane

Biochemical and Biophysical Research Communications 153, 855–861. of pigs. Journal of Nutrition 119, 706–712.

Sandholm, M., Oksanen, H.E., Pesonen, L., 1973. Uptake of selenium by aquatic orga- Woock, S.E., Garrett, W.R., Partin, W.E., Bryson, W.T., 1987. Decreased survival

nisms. Limnology & Oceanography 18, 496–499. and teratogenesis during laboratory selenium exposures to bluegill Lepomis

Santhosh Kumar, B., Kunwar, A., Ahmad, A., Kumbhare, L.B., Jain, V.K., Priyadarsini, macrochirus. Bulletin of Environmental Contamination and Toxicology 39,

K.I., 2009. In vitro radioprotection studies of organoselenium compounds Differ- 998–1005.

ences between mono- and diselenides. Radiation and Environmental Biophysics Xin, G.L., Evenson, J.K., Thompson, K.M., Sunde, R.A., 1995. Glutathione peroxidase

48, 379–384. and phospholipid hydroperoxide glutathione peroxidase are differentially reg-

Sattler, M., Winkler, T., Verma, S., Byrne, C.H., Shrikhande, G., Salgia, R., Grifﬁn, J.D., ulated in rats by dietary selenium. Journal of Nutrition 125, 1438–1446.

1999. Hematopoietic growth factors signal through the formation of reactive Yant, L.J., Ran, Q., Rao, L., Van Remmen, H., Shibatani, T., Belter, J.G., Motta, L.,

oxygen species. Blood 93, 2928–2935. Richardson, A., Prolla, T.A., 2003. The selenoprotein GPX4 is essential for mouse

Schneider, M.J., Fiering, S.N., Pallud, S.E., Parlow, A.F., St Germain, D.L., Galton, development and protects from radiation and oxidative damage insults. Free

V.A., 2001. Targeted disruption of the type 2 selenodeiodinase gene (DIO2) Radical Research Communications 34, 496–502.